Dayton u142

South Alabama u145
Yale o125
Siena o129
S Dakota u143

ODU o118
Harvard o146.5

ODU -12
Baylor -3.5

Cal o143.5
UTEP u132
Belmont +32.5

owned by 3pt shooting on pretty much everything

hate this side of variance

Florida Atlantic o142
Cuse o160.5
Rhode Island -3
Illinois St. -1
Yale +18
Cuse -3
Xavier -10

Marshall -1.5
Irvine +9

Richmond o134
Col St. o128
Davidson u147.5
Nevada u154
Wake o140
Gonzaga u145.5

UNLV u149.5... only got it once :(

Clones more quickly plz.

If you can't tell, generally going to abandon the 5 minute rat race on open with totals to get bigger volume at smaller egdes upon cloning.

Illinois o147
Fresno St. +7

So Miss -11.5

Beat 4 sides at least marginally... and lose all

Get anhilated by the other close, and win it

Eff betting

Florida o131
Idaho u144.5
Niagara o145.5
Clemson o141.5
Clemson -15
Murray St. o133

LBS -1.5
Pacific -4
Drexel +19
So Miss -16
Citadel +14

tulane u134, someone thinks Im wrong, w/e

fml working on a sunday, probably lost more $ in equity from firewallz than my hourly base here

down 0-2 in CBB today, 0-4 steam chasing other sports

grr today

cornell -6 (was greek)
RI -10.5

Webb +24.5

UAB -4.5
SC -1.5
E Mich +5.5
Miss St. -3

scratch WGB side, injury confusion
EKY +10
Austin Peay +24

WGB o138
LVille u143.5
Utah u141
Memphis u143
Temple o130
Texas Tech o135.5
Wofford o144
Oklahoma o143

Pretty sure I've never beat a total by 12 pts.

Least it won. Holla.

CSUN -2.5

Montana St. -1.5 ( -2 for no pinny here :( )

no totals

Weber St. +14.5

Nevada u161

Nor Col -10.5
UAB +2
Miss -6.5
South Alabama -3.5

Wake o143.5
South Carolina o134
Duquesne o129.5
Northwestern u127.5

NMS +9

Looks like Kenpom is forecasting lines again.

Imma start putting out lol March Madness brackets for funzies using his ratings. Probably does more harm than good but seems like people enjoy it.

Also for the sake of karma I should acknowledge I am running white-hot in 2H's this year after the inital whining of 2H run bad continuation from the tail end of last year. Record would be much worse if every bet was 1H.

So Miss -8.5

So Miss u141.5
St. Mary's u160.5

kenpom calculator updated through weekend's games

G Tech o144

W Mich 1H

7-34 FG (20.6%)
1-14 3pt (LOL %)

gg equities

W Mich +17.5
BC o150
Xavier -3.5
UTEP o149
Clemson -21.5
Portland St. -1

No openers because someone has their entire BM bankroll hanging on one game

cough cough

bets were:

GMU -1.5
Air Force +17
Miss St. -5.5
Florida Atlantic +10
Northern Colorado +1

South Alabama o138
Valpo u150
Ball St. u137
Nor Col o146

late to open, probably not touching anything until clones (late tonight), which is I guess is good for following

although maybe a few totals in about an hour

no totals tomorrow

Wafflemegacrushed a few of those totals yesterday. <3 early CBB season. It's probably fine to take a game 2-3 pts off whatever I post for now, as long as lines are going to move 6-8 pts fairly frequently. For BM releases (not late adds with several books offering) we'll move those lines anyway, so it'll take a disagreeing syndicate to get back to that line. And they probably know more than I do anyway. So I dunno.

Took Colorado -2 before crashing, -2.5 still a 53%-54% play imo. It's not massive value. Was considering Montana at pk, maybe if it hits +2. Hard to tell since Loyola is so different this year.

Rhode u140

Harv o137.5
prov o141.5
wisky u137.5
mich u136
uconn u148
San Diego u137

all moved an asston, harv was slimest value, uconn and wisky are 5-6 off IMO

KenPom Calculator

KenPom calculator now over to the right.

Got a couple emails asking me if I knew when Kenpom was going to start showing forecasted spreads/totals for 2010.

1) I don't know.
2) Plz to not be betting them blindly if you <3 bankroll.
3) I created a little Google doc that gives you those lines.

I'm not 100% it's what he would calculate, but I think it's pretty close. The biggest thing I am unsure of is his home advantage, which he once said meant increasing home efficiencies by 1.4% and decreasing road efficiencies by 1.4%.

To use it, you have to spell the name as it appears on KenPom's sheet precisely. not "Mississippi State", but "Mississippi St.". Otherwise the vlookups return errors. You should only need to input the team names to get lines, nothing else. I'll just update it daily with his efficiency sheet and season averages.

Illinois o144.5
San Fran o135

I R <3 Overs

Texas o140
Hall o148

no sides

nice 9-4 today, with some hudge totals moves <3

clemson -6.5
clemson o151
UCLA u140

se mo st. o126

air force u124.5
cal poly u142
ucsb u133.5
cal u152

the citadel -5

Gmu u146.5
Jmu o119
Hofstra u152
Oregon o146.5
Temple o120
Miss st o133
Vcu u144

TN Martin o138 half

Pounded:

TAMU o131
Mahattan +13.5

Have some others, want better linez.

Ships the movement today.

Looks like BM is doing the regular release schedule finally, so I should generally be around for openers if my jubz doesn't get in the way. Can't promise lines will be fresh as horrible totals are rocketing down after a couple minutes now and a whole lot of hitting needs to get done first.

so late to open :(

UW Milwaukee o133.5 half
Loyola Chicago u142.5 half

EKY +9
CoC -10
TN Martin +5.5 (arbz possible)

BYU o136

East Carolina +12
Duke -3 (sigh)
Irvine -2

Jax St. o141
Col St. o132.5

NC St. -4.5

Greek is better at this

Wyoming o136

thats all

Michigan o134.5 half

probably all unless greek releases everything again

wtf massive totals release

BC o140
Norfolk o137.5 (half)

eff got to go

also, full unit spreads bets now, have enough data to at least hit 52.4%+ on ewin there I think

Wright St. u127

thats all

osu -5
luousville u144.5 half

TCU -4
Boise St. -1.5
RI pk

got to totals a little late, not seeing anything obvious

en fuego

OSU -6

Xavier u151.5 (half)
FSU u140 (half)

books seem more willing to release totals when there aren't 65,987,190.4 games to cap

asu u141

thats it, sporadic updating over holliday

Sacred Heart +17.5

First middle evar

almost

Might hedge out of a couple of these

LSU +8
Northern Arizona +7

Virginia +3.5
Illinois -13.5
Wisky pk <- half a half

Cincy o133.5 (smallish)
Kentucky u145.5 (smallish)
Texas o133.5

Middling out Illinois at +15.5 -105

Ew, line crushaments not helping much today so far

Shipzilla <3 the .300 FG%.

Should have kept whacking at it down to 142 ldo.

I'll get the eWin% crap up today if I get bored enough. I make occasional 2H bets but they aren't really followable, only showing game lines. Spreads eWin% will suck, it's going to be a couple weeks before I expect to regularly beat moves there.

It's going to get harder to get my line on openers bets, a lot of these lines are screaming up/down and I have a few people that need to smash it first. I'll post if the live line still has any value when posting in case of hudge moves.

-----

eWin% stuff up

Vandy u149 (now 146, still good, took more)

SC o140.5 ship line move before hitaments

nothing on the others

Sides:
Nevada -2
Temple -4
Buff +4

Totals:
Was informed BP released totals yesterday without BM. Hopefully more volume there. Spreads are half units, totals are typically max bets.

Christ, 0 totals all day? This is bad.

Siena 2H dirtay +4 hedge

Toledo u142

Nova u143 - effing meetings made me miss open #

<3<3<3 taking a 3 to win instead of a 2 to tie Nova

Nothing else on the other sub-handful of totals

Sure would be nice to run better than 50/50 when I wafflecrush lines, and run better than 0/100 when I get wafflecrushed. :(

Also, last season's latter half system where you bet the oppo side of the 2H line that helps my game bet is continuing to crush. Tempted to try it on the Nevada 2H over since I nearly bet that game under imo.

Yeah sure why not, o76 2H. Lox imo.

Butler u128
Texas o138

Not running too hot on the 6 point move crushes :P

UIC +1

GT u129
ISU +100
Iowa o124.5 ty bm

Scratch NW tomorrow, munching some vig

lol just saw UCLA went to to 2OT and over still lost. Nice 5-29 3pt, .471 FT shooting from white dude dead eye shooters. Think I'm staying away from UCLA for a bit, ie this year.

Sweet, my 2H Montezuma's revenge didn't break over the offseason. At least totals are back imo. Please to not go away.

UCLA o140
SDSU u133
Virginia o124
Auburn o130.5

Groan. TY to comment below informing me totals releases will become a rarity in all liklihood. I'll be back in a couple weeks I guess, whenever I have enough data to feel comfortable doing anything with spreads other than chasing steam. I'll probably do a lot more sizing with spreads (no more flat betting tracking) given having half my plays and a high 50's winrate removed.

Boooo

wtf one total releases for like 30 games? Only on Montana for fixz. Back tomorrow, if books bother to actually list totals.

Ship the 5 minutes of free time coinciding with some totals rollout and free wireless internet in hotel. Think I need a few games of game data for spreads, I'm wiffing on movement with bets and near bets. Totals are a little easier since tempos don't fluctuate too wildly year to year under the same coach. Less reliance on blind insight into roster changes.

So. Miss u143.5
UCR u132.5

Stanford -1

Hasn't moved. Hardly anything has moved. Not sure how useful I'll be when lines start flying 5 mins after open, since I generally won't be insta-posting.

Flying to wedding this weekend. Wouldn't expect much of anything. Probably for the best since at the very least I'll know who's getting minutes on each team.

Only one bet for now, some people need more of it, will post later.

OSU u146.5
NC St. u136.5

Christ, that's a lot of free throws. Check ref's phone records imo. I'll post a list of Friday bets sometime later tomorrow, it won't be right at open, but movement seems slow to start off so far.

Cal u141.5

A little weary of Detroit's transfers, who appear to be better offensively, but it's still 7/8 off last season.

Season starts tomorrow <3

Fun with KenPom

So I’m hurriedly preparing for handicapping college basketball season by revamping the model. Somehow the model gets bigger every year. Maybe by 2014 I’ll be making bets influenced by nose size and number of siblings. For ldo reasons, at the beginning of the year stats aren’t the end-all-be-all that they are most of the way through the season. For a lot of teams, last year’s stats are worthless for anything more than some general thumb in the wind guesstimate. Using the returning players isn’t much of a solution either. Individual player valuations are 1) still shitty in basketball 2) even shittier in CBB with little player history and far from static ability levels 3) not all that helpful since player turnover in CBB is ridiculous. Basically, I avoid games early in the year where most of the PT is wearing the same uniform but doesn’t resemble any incarnation of previous years’ teams.

Every year, I also go back through my old kenpom data, since it’s the dysfunctional heart of the model. Kenpom data couldn’t live on its own, but it’s a hell of resource for making your own Frankenstein. As an aside it’s the people that reply to homer message board posts with “kenpom sez, kenpom sez” that help fuel the donklash against it. Exhibit A being the “LOL WHERE IS YOUR KENPOM NOW” internet mob last March Madness season. It’s smarter than they are, but it’s not as smart as $$$ opinions either. Anyway, since I make so many adjustments to the raw kenpom data, I need to check that data for computational changes year to year. Interestingly, I am fairly sure kenpom made some not so immaterial changes to how he adjusts his efficiency stats. Think of it as something like the process of what a 50 point dunkfest over a scrub versus a 2 point squeaker over UConn would translate to in each teams’ aggregated efficiency stats after those games. It’s probably the most important part of CBB because there is such little parity.

The data in the chart below is adjusted by me in several ways which I am not going to go through, so this isn’t pure KenPom prediction & results. It is pure KenPom efficiency formulas, just adjusted to remove some bias, but adjusted in the same manner each year. It shows year by year home cover %’s for all non-neutral games. The spread ranges are at the top.



Kenpom used to have a big problem predicting blowouts. Seeing -21 online and -24 kenpom, and betting accordingly would have been a great way to light some money on fire. But it got a lot better last year. Maybe he has done something like muting the impact of UNC crushing Northwestern Quadrant of Idaho Seminary School, I don’t know. But kenpom might be getting closer to needing less and less adjusting. Just as an FYI, the high cover rates between 0-10 are mostly the result of kenpom not incorporating actual outcome distributions. Team A who (on paper) is 1 point better at home than Team B, will cover -1 more than 50% of the time.

Sweet, I remembered the password.

Inactivity was due to new move/job then internet connection busto then staked for remainder of wnba. I should be around in some capicity for CBB. Lessons learned from 08-09:

1) Bet less march futures because:
-Too much capital
-Everyone sucks at predicting draws
-Prices inflated during the season and deflate at tourny time
2) Leaving money on table with totals volume
3) Bet bigger earlier, CBB is retardo soft in the first month or two

Annoying service outage last couple days in new apartment. GG old building.

Nothing today, will post in mornings so long as internet keeps up.

nothing wednesday

have access again, will post in the morning on game days if there's anything for the day

Apologies for the dearth of plays, got moved to Chicago by my company, finally done with the move and should have regular internet access beyond my iphone in a day or two.

CHI u157 .5u

crazy streak

so much less work than baseball too

CONN -2.5 .5u

CHI -2 1u

nothing tomorrow

good news: model changes finished and now self-updating

more good news: the only bet it would have made today lost

nothing for Sunday

PHO o169.5 .5u

back home Monday, more volume starting that week I'd guess

nothing for Thursday WNBA

PHO u179.5 .5u

Apologies for all the .5u bets - travelling quite a bit and keep getting behind on model updates which I haven't taken the time to automate yet. Betting size and volume will probably pick up in a couple weeks when I'm back.

DET o144.5 .5u

7/3

CHI u145 .5u

WAS u164.5 .5u

7/2

CONN o139.5 .5u

DET o143.5 .5u

nothing for wednesday's game

SAC +7.5 .5u

nothing today either - need to parse out some injury/absence data for a few teams, should have time to do that tomorrow

no wnba bets today

nothing on tomorrow's wnba game

SA o164

travelling today, cant make wnba bets, should have bets for tuesday that morning

6/19

CONN -4

no bet on tonights wnba game

Hmmm, have o/u 160 fair for MIN game based on last year, opened at 161. Blah. No plays right now, need to update CY stats.

running crappy on FG% again in early game, watch SEA and CHI shoot .510 later...

CHI u148

Betting through someone at Pinny now, so there should be some plays a couple days ahead of the games. The other books are annoyingly slow.

6/14

CONN o152

More stupid MIN unders openers please, it's taking combined .500 FG shooting to eek out these one bucket losses on the under.

nothing for Saturday's two games

Doing very simple scaled unit thing now with WNBA, lots of smallish plays come up that'll just be .5u. quite the exact science here.

6/12 WNBA

MIN u158.5 <-- this is off by 6 points using last year efficiency stats, imo. More MIN unders. The teams are not materially different than last year, the totals lines were much lower in their meetings, and all that's changed is MIN is shooting 55% from the field and >50% from 3pt land in their first three games... because that's sustainable. Hell it took that kind of performance to kill the under 2 days ago, and it still only died on the last second shot. You can probably assume I'm auto on the MIN under for a while, I'm riding these to my doom.

CHI -4.5

PHO o169 / .5u

IND u144 / .5u

lol I sweated my first WNBA game today, since I'm bored out of my skull in a random city... and the MIN under lost with 3 straight long jumpers wth 30 seconds left in a meaningless ending. That happens fairly often I guess, probably a little less than 1/10 of the time, but I'm taking that as a sign to never ever do that again.

6/10 WNBA

MIN u161

PHO o166

Groan, I have to bet against Cueto again tomorrow. Taking the 5in line tomorrow assuming it doesn't move much (refuse to take the WAS game line).

9/9 MLB
MIN -107 / 1.1u (game)

PHI +163 / .5u (game)

Basically smallish bets come up on Happ, against Cueto, and a couple random pitchers who are running hot or cold. It's definitely getting to be more trouble than it's worth updating all this crap every day. Every once in a while I guess you find something and go nuts with units, like Beckett returning to form, or Verlander becoming CY again, but it's getting less likely that pitchers will be perceived wildly different from start to start the more we get into the year.

No bet on tomorrow's WNBA game, Detroit is missing 3 of their major players from last year, I don't have a basis to bet on them yet.

Holy line movement, was -6, min +6, was u151 now

Tempted to call it quits on MLB, plays very infrequent now

6/7 WNBA

MIN +8

WAS -4

WAS u158

This feels like betting on competitive Latvian Midget Tossing or something, I'll never watch a second of any of these games and have no clue who anyone is, they're just numbers to me. And I think it's probably easier this way.

If anyone ever comes up with relevant qualitative information like what Ben found about UCLA's uptempo switch midseason, please share, because I'll be damned if I'm reading up on anything but injuries.

Also, the teams are so few and the season so short that any final result this year is going to have a crapload of variance in it, probably enough to throw you in the red at season's end even with getting decent edges regularly. Just so you know.

6/6 WNBA

CONN u155

PHO o162.5

MIN -2

CONN -7.5

WNBA time! Model almost done, bets for Saturday's games late tonight.

Yuck, Lohse early injury and maybe it's possible Cueto has perfected the 250 foot laser beams at left fielders with men on base. STL wouldn't have won either way, but that dude is blessed.

BOS -138 / .5u (game)

ATL -132 / .6u

more 6/3

PHI +110 / .5u (game)

I'm in love with Happ. I typically avoid Young at home with slight model value, but I'm hoping being a FB pitcher even at PETCO isn't going to help all that much with the Phillies.

Wow 5-1-2 today, with some last minute rallies, good error shenanigans, and wins/pushes that would have been losses if I'd done the other inning bet. When it rains it pours, both ways.

SFN -147 / 1.6u

PIT +140 / .4u (game)

BOS +125 / .5u (game)

KCA +149 / .3u

ATL -115 / .4u (game)

OAK +130 / .6u

SLN -116 / .2u

SDN - 155 / .2u

6/3

SLN -120 / 1u (game)

More Cueto babip aments. Last time this moved 20c or something overnight in my favor, then boomeranged right back with the syndicates. I don't think it's nearly as off this time.

love this so much

http://www.mattwietersfacts.com/

I've really got no idea what to do with BAL games right now. PECOTA and ZIPS believe he will instantly knock around AL hitting, actually PECOTA thinks he's already one of the best hitters in the AL. I have no basis to question that other than the fact that there's not really a huge sample size of super touted guys with those kind of numbers. Yeah Braun panned out, immediately, but how many didn't? Kind of hard to put money on the expectation he'd hit 25/30 jacks if we started the season over right now. Maybe I'll just use a 25% PECOTA projection or something so I don't feel like a lemming. If that still gets me a BAL bet, eh, why not.

CLE +172 / .8u (game)

OAK +135 / .4u (game)

PHI -113 / .4u

CHN -140 / 1.1u

Plz to not be getting shutdown by Eric Milton for 5 innings

update:

great getting clutch singled to death all weekend (check SLN push), then 5 straight singles for LAN to start game.

eff baseball and eff babip. sigh. ugh. etc.

SLN -130 / 1.7u

BAL +103 / .4u (game)

TBA -150 / .5u (game)

have them fair for 5inning lines, shrug

I heart the Milwaukee bullpen.

weird fact, MIL 2nd in NL bullpen ERA

http://mlb.mlb.com/stats/sortable_team_stats.jsp?c_id=mlb&baseballScope=NL&statType=2&sitSplit=grp&timeSubFrame=2009&groupByTeam=true&Submit=Submit&timeFrame=1

Might also be time for me to give up Mariners games for my own sanity. Back to streaking their slappy singles in long chains whenever I don't need it.

MIL lineup perfect, CIN missing Phillips. I'm confused. Sigh, arb 2/3+ for 3c loss on MIL, 14c gain on LAA.

SFN -115 / .5u

ATL -120 / .5u

all following are game bets today, low on BJ funds and MB 5in market not always good enough.

Great, would have been a push on 5 inning. Just take the opposite (game/5inn) of whatever I do.

PHI -162 / 1.2u

CLE -113 / .7u

Hmm wonder what's going on with the MIL line. It was arbable for a good profit last night, then swang violently back this morning, so I might have to take a small loss on it. Hope I didn't get lineup screwed (like manager said Braun + Hart get the day off or w/e).

LAN +101 / .3u (FOR GAME)

Can't get a good 5 inning price -> rare deviation from 5 innings for general badassery of LAN bullpen.

add to 5/29

LAA -161 / 3.4u (more arbz)

5/29

MIL +100 / 3.1u (full game)

Arb attempt obv, still fading the cueto HR/FB, BABIP luckfest

Sonnanstine now on my DO NOT WANT list headed up by Andrew Miller. Apparently a 6 spot wasn't enough for 5 innings, err, 3 innings.

SDN -158 / .8u (reduced, lineup mismatch in model)

TBA +102 / .5u

TEX +125 / .3u

lol... since 5 innings only plays there have been two losers and a push that are winners in full game (if Sea holds on). Oh well, the Papelbon blown save more than makes up for it.

I may stick to full game wagers for some of the uber pitchers that go 7-8 routinely (like Greinke, Lincecum yesterday), and live with bullpen coinflipping for the 8th/9th. For scrub on scrub, 5 innings is about perfect.

In other news, Seattle is quickly becoming the Brewers of this year for me. If I bet against them, they're the most clutch/gritty/etc slap hitters with an insane BABIP. When on them, their RISP is horrid and they can't score runners from third with less than 2 out. Oh Mariners, why must you be so perfectly erratic with your slapiness.

SEA -120 / .6u

ARI -160 / .7u (AG out)

SFN - 166 / .8u

CHN -146 .7u

KCA -153 .8u

KCA +127 / .5u

OAK -137 / .3u

HOU +106 / .3u

ATL -132 / .6u

Arb CLE -105 / 1.4u for game at +126 (21c profit)

Arb FLO -105 / .9u for game at -106 (11c loss) <- cantu out, Gload in, TB lineup perfect

No 5 inning bet on CLE (fair line)

FLO +100 / .3u (again, all bets are now 5 inning if not openers)

MB 5 inning volume actually isn't that bad on some of these games for fractional units.

LAN -119 / .5u (no Morales, Napoli)

lol... Paplebon just gave up a 2 run HR to a scrub in the 9th to go down 3-2. Suck it full game bets.

The several cents worse because I can't use MB is totally worth the save in sweating relief meltdowns.

Ship the tilt embracing of full kelly. BOS wins 2-1 through 5. Phillies game marked the end of full game bets at gametime, every line forward that isn't an opener will be 5 inning lines. BOS line was BJ, they'll usually have the best lines. Relievers typically don't hit reliable sample sizes mid year, so it feels like I'm just coin flipping past the 6th inning, especially since reliver use and leverage analysis is very time intensive and I can't get to all of it.

I'll likely also start completely arbing a lot of the openers regardless of lineup, and won't be betting any openers I don't think are going to move 10c for an arb. I'll probably make a corresponding 5 inning bet around gametime the next day if lineups dictate.

Hilarious how bad I run at this shit. Blown multi-run saves everywhere while I can't buy a run past the 6th inning.

I'm probably done if this doesn't hit, baseball is just ridiculous with variance:

BOS -201 / 6.4u (full Kelly, Mets subbed in three shitty hitters, Red Sox lineup perfect, Pelfrey velocity down and not K'ing guys, Beckett's command much better last two games + using FB more)

update: 5 inning BJ line, should be obvious since -201 is nowhere on full game SBR and MB was -185 for game at the time.

PHI +161 / 1.1u

HOU +104 / 1.3u (Kinsler/Young out)

5/24

CLE -105 / 1.4u

FLO -105 / .9u

Nothing tomorrow at first glance. Of course todays loss so far was already annoying. COL misthrow in 8th gives Det an extra run, then they fall 1 run short in their 9th inning rally.

Now LAN (tied 1-1) has one run on like 15 baserunners through 7, I think about 5 huge 3rb base 0 or 1 out fails, two diving/wall smashing catches to save runs, and Torre just put in their worst reliever and walk-a-saurus lefty to pitch to the top of the order - which is righty loaded. Don't know why I bother. Anyway, just win by 4 or 5 on a bunch of dingers one time other team.

wnba yet? plz? Good/lucky unit sizing the only thing keeping ROI in the black right now (negative W-L despite focusing on small-to-medium favorite home teams).

WTF lineups. Molina out for LaRue. Ryan and Stavinoha in also... making -125 exactly fair. Royals lineup is perfect for them. Thanks a ton LaRussa, thanks so much.

Arbing down that whole SLN play for a 3c loss as well. Negating arb profits from SEA yesterday. Ugh this is irritating.

Can't wait to see Dodgers lineup. Kemp and Ethier probably sitting for a couple AAA guys.

COL +137 / .5u

Arbing down SEA... 6c of equity off MB (-116 right now). SEA has probably the worst lineup possible, and I feel like I'm being tarped with regard to Bedard's hamstring injury or something. This should be -130 right now with 100% Bedard, and my stats reconcile with cose for both SP's recent starts, so I dunno.

5/22

SLN -125 / 3.6u

Over-bet, prob arbing. I think this is WTFBBQ off, but then again the SEA -110 hasn't moved much despite losing Abreu for a couple days on a play yesterday, so what do I know.

IIRC, interleague play is sort of like the beginning of the year part-deux in terms of line moves, so there might be a few more gigantic arbing attempts.

Adding:

LAN - 125 / 1.2u (might arb, not sure about Abreu tomorrow, or what rotated DH won't get to play)

5/21

SEA -110 / 1.8u

another wtf line, imo, slight overbet - might arb

watch it close at -105 now

added: BOS -165 / .8u

LOL this is really getting comical, I can't lose a bet without shenanigans. Up on the day since PHI won - in totally standard fashion. CLE and TBA lost, again in hilarious bad beat fashion. TBA 2B boots a ball in the 11th with two out, which allows Holliday to come up and hit a 3 run jack. Then Kerry Wood gives up 4 runs to the Royals in the 9th with 1 out and none on. Just bizarre how I lose games. Usually the pen too. I swear, I'm actually reverting to 5 inning bets on anything that isn't loltastic value with open (since line will correct before 5 inning lines get posted). I don't model bullpen usage so I even have justification for it!

Ah Votto out - price is fair right now, no arbing.

Nothing big for Wednesday

PHI +100 is <3 <3 <3, but I think it may have boomeranged too far the other direction (-134 on MB). If you're into that sort of thing, arbing out here is probably going to be a better price than at close, unless I don't know something about the lineups today.

rapid course correction imo, 3 in losses in a row, and FLO closing crush/1-0 lead is going to get rained out.

.4u more on LAA at -113 (Beltre/Cedeno swap)

Beat me senseless if I ever bet WAS anything other than 5 innings again.

5/19

PHI +100 / 1.3u

Phils line is perplexing. Cueto runs like God with BABIP and this line, given normal lineups, says Cueto = ~Hamels. Well, I get a no bet if I plugged in Hamels vs. Hamels anyway. Waiting on BJ for possible others.

Update: had to bet rest through iphone on Greek. Here's the other small adds, they all moved slightly.

CLE -135 .3u

WAS -133 .3u - arbing out, might bet 5 inning

TBA - 180 .4u

Back to normalcy. Winning streak stopped with a couple one run losses: one despite my side throwing a 2 hitter, and the other because of a bottom of the 9th error. Small units, so w/e, but I'm pretty damn awesome at picking the weird losers rather than the standard 3 run loss shellackings. Kinda wish I'd get the latter, less sweating. Here's to losing by several runs and being out of it in the 6th inning tomorrow.

Back from my trip, so opener betting starts again tomorrow, assuming there's anything to bet.

5/19

WAS -110 / .3u

FLO +124 / .3u

LAA -102 / .2u

Bos -132 .4u

sln .3u / -1

Micro plays for 5/16:

MIL -105 / .5u

that's all

Obv Chris Young on the road is an odds don't matter system.

Nothing tomorrow, won't be around for open the next two days, but I'm going to start making some micro bets before limits go up overnight. Sitting on your thumbs waiting for big plays is boring, and I'm probably passing up value, though I haven't kept track of movements vs. close with "leans" or whatever you want to call marginal model value.

I Run Like God

7 game win streak assuming Oakland doesn't blow a 9 run lead in 2 innings to the Royals. I've already marked it a W, which might be asking for an epic bad beat here. Anyway, nice reversal of fortunes this season. Units updated/corrected to reflect all the early season arbing crap, which is basically over since beating close isn't a given anymore on most of the plays.

Groan. Chris Young on the road opened fairly (imo). So much for that angle.

Nothing tomorrow.

Meh Pinny out, and not giving a better price than Greek/BM.

5/12

OAK -125 / 1u

One smallish play for 5/12, waiting on BJ for it. It'll be up when BJ opens. Not expecting it to beat close much, hopefuly 5-10c.

Nothing tomorrow - only four games.

No big openers tomorrow, maybe something small when BJ rolls out

Ok NOW I'm using up some of that late game karma, 6 in a row. Crushed close, so I don't feel so bad. The edges are getting smaller by the day - I'm forgoeing most of the micro plays, but I might be forced into taking a few per day pretty soon, since practically every game not involving a rookie pitcher is opening at fair value.

Nothing for tomorrow, but Saturday is a big day for random manager-granted days off, so maybe a few gametime bets.

Yay run good, 5 in a row, about 6% ROI with all bets now. So far, I haven't really used up any WTF-are-you-kidding-me game-ending suckout karma from some heartbreakers earlier, so hopefully there's more to come.

Still generally beating close on openers, but it's more like 10c instead of 20-30c now. Probably approaching the point where there's not much sense to bet large spread openers without a newbie pitching (like today's LAA game, and tomorrow's MIL game). And... that's going to coincide with WNBA pretty nicely. Go beaverball.

The instant value of Manny apparently:

Kind of interesting to see the importance (approx) of a player over his replacement. IMO this should have dropped 19c, not 13c, and it's dropped more since but that's likely not due to anything Manny related. A lot of my gametime bets are reactions to lineup changes that I don't think have moved the line enough - perhaps my variation between player valuation is wider than most, but I can't really think of a reason why right now. No real point to the post, just rambling.

5/8

MIL -110 / 1.9u

5/7

LAA -154 / 1.2u <-corrected, wrong catchers

just changed to -160, adjust Kelly by win probablity accordingly

Yay on finally (Jays game) luckboxing huge inning down a few runs, although that was made necessary by various errors/semi-errors that accounted for half CLE's runs. Mixed emotions.

Go Timmy.

In positive unit territory again, plz to be staying here.

5/6

Took SLN -130 on Greek, .6u, doubt it moves much, just won't be around tomorrow. Still there on pinny if you have it.

SFN -111 / 1.9u

Cubs lineup is a total joke today

No Soriano, Bradley, Soto

Groan, hey Cleveland sit a starter or two. At least don't put up the best possible lineup. TOR is sitting regular catcher for a practical pitcher-equivalent at the plate, which is brutal.

Arbed down (1c loss) to .9u

5/5

TOR -110 / 1.8u (go Cecil)

Chris Young on the Road - My First Stand

Note: That stuff you can get from fangraphs. How to get detailed pitchfx data: http://fastballs.wordpress.com/2007/08/23/how-to-build-a-pitch-database/

Well, here's yesterday's bet in partial data form. Something happened to Chris Young last year besides getting horrifically slammed in the face with a ball. Fastball velocity is down, and it's persisting. It also has less horizontal movement. Probably as a fairly direct consequence, his line drive percentage jumped to 24.9%, which likely also involved a little bit of bad luck. BABIP is used to eliminate luck using GD/FB/LD rates, but luck still exists in the breakdown of those rate stats themselves. Anyway, the drop from 90 to 87/86 is significant, very significant. He ran ridiculously hot last year to end the season with respectable stats (check the eBABIP). He is also an extreme fly ball pitcher in San Diego - AND ran very hot with HRs there.

If I had to guess, the marked increase in the use of his changeup (20% this year) is probably directly related to the speed issues with his fastball. Obviously, a good change improves your fastball, but this change - whatever he's done different - has no horizontal movement, it's gone. It's also only 8 mph slower than his fastball, and the speed gap between a change and a fastball is very important. Obviously the guy still has a unique release point, being 6'10 and throwing over the top, so he's not Barry Zito - but I'm also of the opinion that nearly all the weighted-means and comparable projections are optimistic, and expect more or less last year's performance, which was lucky to be as average as it was. Game lines (like last night), imo, are reflecting the common projections.

Of course Young pitched decent last night, but avoided a couple high run expectation situations. Anyway, this might be slightly psychotic, but I'm going to betting against the market close with Chris Young on the road for the near future. Maybe at home too, but he's perfected the deep fly ball out there, so I'm a little wary that hitters aren't crushing 440 foot bombs too often, and hit a lot of HRs via the 340 foot variety. I'm a huge believer in market equity, but I think it's possible to pick your spots from time to time, especially early in the year, so long as you aren't betting half the games per day at close. MLB's more efficient than CBB, but see my post regarding early year equity/efficient close on that subject.

So obv Loretta can't pull a Lopez. Now Mota is in the game. I'm going to lose with 95% confidence.

OR NOT. FINALLY.

It's weird to say, but I can't wait for the WNBA in June. Betting women feels like coin flipping, but baseball is way more tilty.

I run like ass

Bottom 6th, 1-1 game.

- A. Ethier singled to right

- J. Loney doubled to left, A. Ethier to third

- R. Martin grounded out to shortstop

- M. Kemp lined into double play shortstop to third, A. Ethier out at third

SD is retardedly playing all infielders in with a power hitter up and 0/1 out. Balls basically have be hit right at them, or they're hits, and that scores two runners. And of course Kemp's ball is only caught because the dude is right on the grass and the ball is hit directly at his shoes. Amazing. Another horrible coaching move, but this time by the other team, ironically owns me through luckboxing.

I think the only appropriate way for this game to be resolved is probably david eckstein grounding out to third, but two throwing errors allow him to make it all the way around to home. And probably a Manny 403 foot out to dead center to end it.

I'm getting the ominous feeling LAN +600 is about fair right now, tie game, superior lineup, and home field be damned.

LAN -160 / 3.1u

Should be back to betting openers tomorrow, been busy. There hasn't really been much anyway, arbing days are probably over and I ran like ass during it. Lame. I'm starting to alter projections based on YTD rate stats, slightly, so probably more plays per day as I take an opinion on players projection paths (using pitch fx for pitchers to identify material changes in velocity and movement).

Just got home - see SEA and OAK tied in bottom of 9th (OAK blew 5 run lead obv), 1st and 3rd, two out, Branyan up...

And Oakland is walking him. FFS, a HR = a single here, you lose if either happens. Lopez is about .030 better against righties anyway. And then combine higher odds of hit + odds of walk to Lopez (on deck) and we're >>>>>>>> odds of Branyan hit. DO NOT WALK THE BASES LOADED. Hell you already have a force play at 2nd. This is madness. I bet it ends horribly.

Mid at-bat update: Lopez has swung at about 5 balls in a row on 2-2. Wow.

Final update: I may not run well, but I predict pretty good. I'm getting a little annoyed losing games in large part because of fat overpaid 50 year olds, rather than the actual players and stuff.

5/1

OAK -102 2.6u

cant bet open today, will be back later

Christ I run like ass when I murder close. Practically every big win so far would be a win at -1.5 lines, and I'm hugely negative in one run games. NYN game is extra annoying: FLO had like 3 clutch hits, NYN has like 4-5 rocket grounders with diving stops to prevent 4-5 runs. 

Some pretty horrid managerial decisions today so far too: NYN pinch hits a good fastball hitter (with bases loaded two out in 9th) against a fastball pitcher. He replaces the guy with an absolutely atrocious hitter because that one has more "speed" or something. /life. BAL was pretty bad too. Had a chance to come back down down 1 run in 8th with a 3-1 count, 1 out, Markakis on 1st... and the manager sends Markakis to get gunned at 2nd on a hit-and-run. Markakis is not fast. The hitter at the plate whiffs a lot. Grr.

There's got to be a site that attempts to quantify managerial decision making (like bunts way too much, pinch hits the worst players, loves hit and runs with strikeout artists, brings in his worst relievers with game close), because I really need to incorporate that.

also: had bet on bos... now no youk/drew, arb out.

adds:

NYN -175 / 1.6u (no Uggla, Ramirez for FLO)

BOS -127 / 1.7u (arbed out - lineup changes)

I can't figure out the Yankees this year. The Hughes play only finished 4c better than mid-market. Tomorrows BAL and TBA lines are crushing, but my NYA line sucks. My BOS bet with Masterson v Pettitte moved heavily towards NYA when I had huge value on Boston. I suppose the answers are probably Hughes and Masterson are projected optimistically, and this absolutely atrociously projected Detroit rookie tomorrow is supposed to be decent. It really seems like most plays lately are coming because of a young pitcher, where PECOTA has to fill in for lack of history. Either way, it's a fairly safe bet I'll be on Masterson and Hughes' next open - but I really can't guage the market well enough to overbet for arbs.

4/29 (all BJ)

BAL -115 / 2.1u (also took -120 on BM, arbed half)

NYA -140 / 5.1u (1.1 open, added 4u at -140 on MB on gameday (full faith leap with Pecota))

TBA -115 / 1.2u (1u now, arbed .2u for lineup construction)

4/28

NYA +104 / .8u (BM)

Meh, arbing at -147 and taking a loss on CHN. The lineups are going to be as disadvantageous as possible. Course both arbs yesterday won...

Tomorrow's plays shortly.

4/27

CHN +130 / .7u (more if Ramirez or Lee aren't still out)

I'll try to put up some general mlb luck stuff tomorrow. Pythag records are pretty easy to find, but there's a lot of luck behind the actual runs scored too.

add ons:

-no positive lineup changes

-Arbing out of LAA at SEA +140. I'd bet small on SEA right now if I had to bet again. LAA lineup is gutted, got 22c of positive movement (took LAA -118 at open), so taking the money and running imo.

-Also arbing out of COL +121 at LAN +104. The three starters caused that movement, but Helton's also out for COL, and again I'd bet LAN at a very small edge if I had to. Taking the 25c arb and running again imo.

-SDN had big equity but Peavy is just pitching like crap - the command isn't there to start the season. Walk rate up, watched him struggle pretty badly a few times in the last starts to grove a strike in an automatic take situation... Probably staying off his next start - same with CC. IIRC Peavy had some small issue in spring training, can't remember what it was.

-Groan and now Yankees too. Wrong way on Yankees movement late, market knows something I don't I guess - which is extra strange considering Berroa is replacing Damon and Gardner batting leadoff.

I should probably give up this handicapping thing and just start a Lehman Brothers-esque humongous leveraging arb-log (ie: 13 units on COL last night, arb it all out before first pitch). Of course, that'll probably be when I start running good (or even at expectation) with all that equity. The occasional internet or matchbook server outages before games might make me suicidal though.

COL +121 / .6u

Manny, Furcal, Blake sitting tomorrow.

Fun with Brewers

Christ, MIL blows 2 run lead in the 8th. Serves me right I guess.

Hart lead off the 9th with a double that was like 5 feet short of a dinger, but I'm sure they'll find a way to not score even with Braun/Fielder coming up, run expectancy matrix be damned.

Update: Fielder 2R HR! Ship the jinx!

Update part 2: Shit, there's still the bottom of the 9th... with Berkman/Lee.

Update part 3: Fuck. You. Brewers.

Update part 4: Hope?

Update part 5: lol MIL takes 1 run lead in 11th, but can't score any more with bases loaded and 0 out. That's gotta be a precursor to disaster.

Update part 6: Mike DeFelice for President.

CLE -117 / .4u

SDN -120 / 1.1u <- oops, already posted in opener

OAK +107 / .6u

BOS -135 / .8u

4/26

BOS -132 / 2.5u

SDN -165 / 1.6u

CLE -116 / 1.5u

LAA -118 / .6u

How do you avoid Brewers bullpen blowups?

...bet on starting pitchers who throw complete games.

Still getting some decent movement off open, eROI stabilizing at 4.7%, but giant overbetting arbing plays are getting real rare. I balked on KCA (Greinke) because I didn't have multiple projection sources for that Detroit rookie, whoops. Trust in PECOTA I guess.

4/25

ARI -115 / 1u

MIL +122 / .7u (god help me)

SDN -120 / 1.1u

TBA +100 .8u

4/24

MIL -115 / 1.1u (BM), prob won't move much

Well, somehow Zito, Looper, and Pineiro combined for 21 innings of 2 run ball, BJ Ryan blew a 3 run save, and I ended up positive on the day. The dramatic moves off open are starting to slow down quite a bit. Today is a good example - Billingsley finally opened in fair territory. I might still bet that game, but it didn't have 30c off open like usual. So, probably less betting huge cent spreads at open, and more gameday betting on matchbook, bj and dimes.

TOR -137 / 2.7u

Jones in for Hamilton for TEX, line moving wrong way.

TBA -104 / 2.6u

DET +102 / 2.1u

nothing big to arb later for openers

4/23

TOR +106 / .8u

4/22 adds

Not counted as openers - don't expect to beat close much. Previously, I was sort of shutgun modeling - spitting out lines using many different projections and two different run estimators, then betting when it all agreed on a play. I'm removing outliers now, and putting more faith in PECOTA/Chone, the better projection systems. It's going to cause quite a few more plays early on, but I still have the market data to get a good sense for whether to take something at open which may move quite a bit, or wait and concede I won't beat close by much.

PHI -125 / 1.1u

NYN -102 / 2.4u

Sweet, 27% below expectation in ROI

Baseball is so ****ing frustrating. Demolish close and get raped. 4.7% eROI on baseball moneylines is ridiculous value territory, and I'm getting owned.

On the three large plays, I get spotted 3 runs, 2 runs, and 1 run in the top of the first inning, and I'm going to lose each one. SFN and TBA are still going, but I'm not holding my breath. Anyway, go look at a win expectancy chart, multiply those together, then enter some raffles with those odds, because you'll have a decent shot at winning a big screen TV if you can keep this up. And each loss has a nice little WTF ARE U SERIOUS element to it.

LAN: Down 3 in 9th, two on no out and Manny hits a ball 6 feet short of the fence.

SFN: Edgar Renteria of all people hitting grannies off JAKE PEAVY.

TBA: Worst of all. SEA got the lead with a three run inning by hitting two rbi triples that are both routine deep fly balls and easily get caught by a guy at normal depth, but they've got Kapler playing little league distance in center for some insane reason. Nice little 375 foot out in the 8th that would have tied the game if it had an extra foot, too.

Anyway, the days of a few plays are over. I'm being too cautious this early, and variance is just going to own my ass with a couple plays per day. Especially when things get ultra sharp in mid-season. A lot of the plays are probably just going to be matchbook lines early the day of the game. I'll still bet some games the day before if I expect movement so I can gobble up some useless useless equity.

Update: And Arizona blows a 3 run lead in the 6th, giving up 4 runs. Yeah. FML.

4/22
SDN -110 .5u

Nothing big, maybe a few more later.

Also, fyi, my unit size on TBA is probably too big given that it hasn't moved. Probably arbing some of it for a very small loss, but LAN and SDN have huge equity - not arbing those plays.

Forgot to add:

SDN -105 / 1.8u

4/21
Some small over-betting on some of the larger units to leave buffer room to arb for lineup surprises, but these aren't 2x Kelly arbing attempts.

LAN -130 / 2.8u
ARI -115 / .8u
TBA +100 / 1.9u, -105 / .8u
TEX +200 / .2u

Groan @ bullpens today, lights out for TOR and MIN against OAK and LAA, not so much for CLE against NYA. TEX went right, but smallest to-win of course. I might start betting MIL, WAS, and a few other teams on 5-inning lines. Bullpens are difficult, since it takes a long long time for relievers to hit sample sizes that give me statistical justification to question projections.

Nothing huge tomorrow. Probably wait for lineups.

Yeah, not sure why I bother with small plays early, I'd probably get a better price waiting until gametime. Checking lineups now, which might cause me to make a few more plays or arb out of current ones.

ATL -107 .2u
BAL +180 .1u
Arb OAK play for loss down to .4u (No Nomar + Cust + lefty = useless)

Big equity today, and finally back in positive territory. I sat in the first row behind the Baltimore dugout at Fenway, so I'd like to think my heckling of Eaton added an extra cent or two of equity :). Anyway, always nice when you effectively get paid to go to the game.

Also, rooting for the Red Sox in person is painful. I feel pretty dirty right now.

I can't remember if this was the situation last year, but generally the big equity from line moves are coming from openers underestimating great pitchers and overestimating horrible pitchers early on. Eaton shouldn't have a job, and Billingsly is superstar territory. It's almost like with the exception of accepted superstars (Lincecum, Santana, etc), openers are assuming a floor and ceiling just before the extremes for most all other pitchers.

Posting from Fenway

YOUUUUUUUK

4/19

LAA +130 / .6u

OAK +110 / .7u

CLE +220 / .3u

TEX -115 / .3u

Not terribly confident these all beat close, so no arb attempt today.

There a couple medium plays tomorrow - waiting on Pinny and BJ.

Broxton pitching 1.2 innings tonight, sigh. He'll probably be unavailable tomorrow. That's worth several cents, so I'd arb now at 152-155 if you got -150 and took more than you want. It also reduces the Kelly stake to less than half the posted amount.

Christ, blown late inning leads on every play today, including a 5 run lead in the 8th. Standard.

Also, I don't think I can bet MIL or WAS games anymore, half-seriously. They're the team versions of Javier Vazquez. They're composed of a larger than normal collection of players who don't come close to their projected numbers. So, I'm not sure how to value them. MIL beat close slightly, only probably because lots of people think Hernandez sucks. WAS didn't beat close. Anyway, that reasoning is the rational part.

The less rational part is that they also lose games in tilty ways, which makes subsequent shitty bets more likely. Today was a pretty good example. WAS blows a save by a lofty HR that clears the fence by a couple feet, then get behind in the 10th on a rinky-dink roller between 3rd and short, then Dukes crushes a ball (leading off down 3-2 in the 10th) about 6 feet too low to be a HR, and the next few guys proceed to watch balls down the middle like a bunch of tourists. Dukes never gets further than 2nd. And of course the MIL bullpen blew their game, complete with an ending that involved Louis Castillo luckboxing a game winning ground ball slapper with 2 outs that was too slow to ever throw him out on.

So yeah, probably bye MIL and WAS for now, I don't trust the projections of a solid handful of starting position players and relievers, and sweating your games takes years off my life.

4/18
Again, arbing half on big units unless there are lineup surprises. Arb gains/losses in ROI based on closing matchbook price.

LAN -150 / 4.4u (arb)

BOS -207 / 3.2u (no arb)

TEX -113 / .2u

Groan - Guerrero out indefinitely. Arbing the LAA play down to .2u using Matchbook.

4/17
LAA +110 / .2u
MIL +125 / .5u
WAS +115 / .3u

If it weren't for the Giants, this would be ugly so far. I can't really hope to keep up the 4% eROI too (4% is roughly beating Pinny vig-free line by 10-15c every game), so it'd be nice to run good while I still have the confidence to bet large edges.

CLE +180 is the only thing I'd bet now - but I read somewhere that Lee's fastball is off and he's using it less, so I don't have a ton of confidence in the projections for him.

Nothing yet for tomorrow. Also, I'll break ROI into openers and game-time bets. I'm obviously not going to beat close betting 30 mins before gametime, so I won't bother with eROI for gametime bets.

Christ, already running bad. Simon (Bal SP, one of the worst in the league) gets injured in the 2nd inning, so Texas gets shutdown by a much better bullpen. Now Harden is pitching like Zito and Marquis is... good now? Also, if I make mention of planning to arb a large unit size (~3+), generally I'm overbetting 2x and arbing half, so about half the posted play is the Kelly stake.

Betting on posted lineups now. This part is pure capping I guess, but I've never thought lines adjusted to lineup changes enough (Giambi/Nomar both out for A's).

ARI -106 / .6u (MB)
BOS -104 / 2.4u (MB - sbr lag)

Again, prob arbing some of this:

4/15
All Greek
CHN -200 / 3.1u
LAN -140 / 1.6u
LAA +100 / 1.6u

adding later:
CLE -101 / .3u
HOU +112 / .2u
TOR +126 / .2u

Whoops, overthought the Zona game, -131/+123 now. All my maybes have a little equity, might start making 3/4 plays per day early on (this will end by June probably), but still very top heavy on the one or two games I think are way off, if they keep presenting themselves.

TEX -125 (Greek early) and -128 (BJ now) / 3.5u (overbet counting on movement, may arb most)
MIL -130 / .5u

On the fence about Ari +141. Numbers say yes, but Scherzer just came off DL, was apparently wild in his minor league start, and Carpenter's forecasted numbers are conservative if he's the Messiah again...

Also, the Giants are this year's anti-Tampa Bay, imo. That lineup is still total trash, and I think an improvement has been built into the openers thus far.

Groan Yankees, 8th inning two out no on, 1 out from from Mariano and poof.

LAN -150 / 2.9u

4/12

NYA -133 / .7u

NYN -119 / .3u

To give a better sense of bet scaling: I'm betting half-kelly, 1u = 1% of BR, so 2u+ plays will be fairly rare, and fractional unit plays common. Double/triple my posted units and you're probably closer to most cappers who probably have 1u = .5% of BR or less. While necessary for scaling, units are a semi-useless way of evaluating a record, so I'll keep results in ROI terms mostly.

Model done

Model complete, handedness/defense/park factors finished. PHI and KCA would not have been a bet :(. Real Kelly units, bet sizing, eROI tracking starts now:

4/11
SDN -142 / 1.4u

Added note: there usually won't be more than a couple plays per day. MLB is much more efficient than CBB and I think too many people employ a shotgun approach. It's hard to compare, since hardly anyone else tracks eROI, which is a much more important indicator in well understood major sports with high limits.

Still on course to be done on Friday, have some cleanup work on handedness to complete. Floating a few small anyway: WAS +190, SDN +118, KCA +162, PHI -132

Back

Well the final four could not have gone any shittier for futures. I hedged out a few units on a couple teams, but we got left with UNC and Michigan St. in the final - two teams whose futures' prices were nearly constantly horrible. I had damn near my entire bankroll during the FF trying to get enough UNC ML to bleed tiny amounts of equity out of remaining futures.

MLB betting will begin on Friday, I need to load a whole bunch of roster updates and start quadruple checking the calculations. I found the MLB betting log from last year, and calculated an ROI and eROI. 3.2% was the eROI, actual ROI was 7.6%, and most of it came during the beginning and end of the year. It should be noted that performance was generally against Greek openers, not WA game day lines. I moved the Greek openers myself 5c sometimes a few minutes after open, so it may be hard to get some lines immediately, but they rebounded back at some point frequently. Also, matchbook is indispensable for baseball. Nearly every bet will be Greek or MB, with some BJ/5Dimes/BM once I get better funded at those books.

Since I actually enjoy baseball, and it's a lot easier to parse information than basketball, I'll probably be doing some general semi-betting related SABR type posts to make this more than a bet-logging blog.

WNBA starts mid-May.

On vacation on remote island, probably no more cbb updates. Wnba/MLB stuff starting next week. I've never bet wnba, MLB roi was around 7-8 % over the course of the season, but higher to start. MLB record tracking will include bet sizing, since almost all bets are on the moneyline. I'll keep an eROI versus close as well.

Ew, nothing like 18 points in the final minute on your unders. New market bracket:

Market Odds

My best guess using Pinny/MB futures and game lines.

UTEP u132

UConn u149.5

Bodog also has some to win futures pretty out of line with Pinny.

ND -3

SDSU -3.5

Pacific +5.5

Baseball model is finished. It's not any different than last year except it's not a total PECOTA whore this time. Season starts in a couple weeks.

Baseball is a lot like CBB, or just about any sport. Openers for the first month or two of the season can be way off. It took books weeks to adjust to consistent early Rays betting last year, although they're probably an extreme example. If you bet it, start early, imo.

Bodog Still Sucks at Props

2000 simulations using market bracket:

2000 simulations using Pinny's FF futures (yes/no vig free):

Error range of about 1.5% again for the big %s.

0 has some value. 1 at 12/1 is just a total linesetting fail. $75 max. The BE and B12 props have no value.

Penn St. +11.5

Greek has some tourny futures up, including what I think are fair ~ prices on Michigan St. FF (+325), Connecticut to win (+665), and UNC FF/to make final (-155/+155). Really only useful if you need to hedge a small amount and have most of the dogs still in the race.

Duke (+1615), Memphis (+805), and Kansas (+3250) also all have about fair odds to win on Greek if you need to hedge there.

I don't think anything but Michigan St. FF has much value on it's own.

Bracket updated to reflect changing MLs. I also took UConn/Purdue down a notch, that half of the bracket got a little too high of a general uptick during reconciliation.

March Madness Week to Date

So I spent some time earlier talking about how donkarific some of the openers are for the first weekend of March Madness. Capping results:

Spreads: 6-5 (eWin%: 55.4%)
Totals: 3-3 (eWin%: 55.6%)

Running a half-win below expectation unfortunately, but March didn't get any sharper this year. Sweet Sixteen and beyond will probably be more difficult.

And below is the YTD, rolling 100 plays to smooth out the variance. Mostly a whole lot of run good, but some of that is CBB being inefficient at the start of the year. Lately, a whole lot of run bad, but equity has actually picked up a little from midway through.

Taking Xavier +7 and Zaga +8.5 small.

Market bracket is up. It's facially strange how odds get worse for some teams that win games, but a lot of that early equity is wrapped in top teams' upset potential. 1/2/3 seeds aren't losing. The odds of them winning all 24 games was really small.

Remember that this bracket is done on a snapshot basis. The Pinnacle moneylines are obviously very fresh, so I'll redo this in a couple days when the lines sharpen a bit. Or, if some gigantic 2pt movement occurs or something.

New market bracket on the right. Everything again adjusted to match current Pinnacle lines if applicable. Obviously I'm guessing at some lines that aren't out yet (UNC/Zaga, Duke/Nova).

Weeeee swongs. 3 and 4 points of equity on the total, 1.5 points on the sides... and 0-3 with two late game screwjobs.

More losing plays with big equity coming in a bit, going to add to the last post.

Also, I'll update the market bracket in a bit through today's games.

Yay run like god yesterday in late game. Good equity, 5-1 capping plays, fail on steam chasing.

Ship bodog 6/7 midmajor too

ASU u139.5
Oakland o138
Richmond -4
Kansas -7.5

Just Pick Winners?

Bleh, a gross 6-11 past three days with average expected win% of 54.2%

Equity is worthless imo

UConn u139.5

UNC o155

Zaga -10

Penn St. -1

Just going through Bodog's props real quick. Limits are $150, so nothing to get too excited about. Greek had some great last longers (Clemson v. BC -130, Purdue vs. FSU -130, etc) but those pretty much insta-move when hit, so I can't post those up here really. They look about fair for the most part right now.

2000 simulations using Pinnacle moneylines:

Error range is ~1.5% for the higher odds wins (5-8). Betting 6 and 7 on bodog has some value.

March Madness Bracket Odds

March madness bracket odds are over on the right finally, and it's my best estimation of a true market bracket. All the first round games' win percentages should roughly match up with Pinny as of a couple hours ago. Again, mostly I just tweaked high level efficiency stats to get as close as possible to deriving the 1st round line, and past two games' lines. Actual game lines could easily be off by a point or two in the next round, there's really no way to hit it completely on the head with a high level formula. But, it hopefully tempers some of the more major over/under expectations of KenPom vs. market. Using spread -> ML conversions also avoids the log5 underrating big faves problem.

One major caveat: teams are static throughout the entire tournament in this analysis. Fairly obviously, a 10 seed making a run to the Final Four is probably going to be less of a dog in the Final Four than they'd be against that same team right now. A 1 seed in the FF is more likely to have simply met expectations, and won't see a similar market boost. So, favorites' odds are probably somewhat overstated, underdogs' odds somewhat understated. North Carolina is very overstated if Lawson is seriously hurt. This presumes healthy Lawson.

Got too busy to finish market bracket today, will be up by end of tomorrow. I suck etc, it's really not all that different than pom. You won't be dashing off to some line that looks meh right now.

Bodog props (very low maxes):

Will a #1 ranked seed (Louisville, Pittsburgh, Connecticut, North Carolina) win the March Madness 2009 Tournament?

-No +190 has sick value (+100 fair)

Totals:
UCLA o135 (small)
Cuse u134.5 (big)
ASU u124

WVA u129

Memphis u132.5

Other tournaments (all small):

UNLV +7.5

Washington St. +4

Troy -1

Kentucky u135

Washington St. u120

Pacific u128.5

K St. u138

Vermont o145.5

I'm out to Vegas in the morning, so I'll miss totals if they come out then. I'd expect a market odds bracket to be up fairly late tomorrow night. So long as the books get their props out, I'll be done hitting most of what I need to by then.

Duke +2750 on Greek is off (not finished, but prob 17-1 by market).
So is Louisville last longer vs. UConn +120 and Louisville vs. UNC +200 on Greek. Those lines should be something more like -150 and +120, but again, not quite finished with market odds. UConn's draw is horrible, Louisville's is the easiest path of the #1s.

I'm almost done making a market odds bracket for the real thing. It'll be up late tomorrow. There's probably already 80 zillion KenPom log5 brackets out there if you're jonesing.

Openers:
Cal pk
Minnesota +4.5
American +18.5
Morgan St. +17.5
Robert Morris +17.5
Wisconsin +3 (small)
Ohio St. -3 (small)
Michigan +6 (small)

Wash -4.5 (small)

small steam chase gogogog:

Marquette -4.5

WV -9

That's it for now, probably.

Probably no plays tomorrow. I'm leaving for the day very early, and I have no clue when these totals for the 10AM games are going to show up.

The futures/props have started out better than ever. I had max amounts of:

Zaga +100
Kent St. +1200
Akron +800, +650
Temple +550
ASU +650
WV +1500
UNC +150 (bad in retrospect because of Lawson's injury)

Ended up somewhere around 6u, and wouldv'e been a little more if I hadn't hedged so damn much, but EG is EG.

I don't know how quickly March props will come up once brackets are announced, but it's criminal not to be well funded at several sites for this. I'll throw up anything I spot as usual.

Nice! Psyched out Gamblor apparently, since the 2H sides that helped the game bets went 3-1 today.

3-1 on the day, one total flat, one a couple points of equity, big equity on Missouri side, and not much movement on the FSU side. The FSU line seemed like an anomaly. +9 is about two points worse than I'd expect with Lawson, and that team seems a hell of lot worse than two points without the most efficient player in the nation and virtually nothing talented to replace him. Market knows better I guess.

3/15
No sides, totals are late.

3/14 (bet the 2H opposite side obv)

FSU +9

Binghampton o134

Late morning lines:

Missouri -4 (full unit)

Morgan St. o133

Bracketologists

The Bracket Project keeps track of a crapload of internet and TV bracketologists, and updates their selections daily. Here's a link to a giant matrix of those various bracketologists' predictions as of last night.

http://bracketproject.50webs.com/matrix.htm

I don't know if there's much wisdom in the aggregate, since a large chunk of these guys must have a learning disability. I'll bet my grandmother that Oakland and Weber St. don't make the tournament. But, remove the whackos and there might be something to it.

I'm not entirely sure if their bracket is just an aggregate of everyone they survey, or if they do their own, but it looks much better than B101. They also have Memphis as a #1 seed, which I'm not sure about, but I love seeing it on paper given how much of their futures I have. Bracket Project's aggregate bracket, or whatever it is, is here:

http://bracketproject.50webs.com/

That bracket will be over on the right shortly. Ultimately the seeding and regions aren't huge deal-breakers for futures, provided you have some diversification. But, it's helpful to have at least a couple scenarios. That way you know whether to pound 40-1 on the strip when 20-1 looks fair in one bracket, or if the value is just a mirage because their draw is best case scenario.

Deja Vu x 13,546

2-3 today on the big bets made last night, with 2-0 totals, 0-3 sides, still ended up slightly because of higher bets on totals, but this. is. getting. absurd.

Guess how you would have done betting those 2H line that helped the big game bets. Yep, 1-4 again. This is nuts, I'm approaching lottery ticket territory with the 2H screwage over the past could hundred bets. If it goes on any longer I'm just auto-hedging everything at the half on the assumption I've stumbled ass backwards into some miraculous method of predicting 2H losers at a high 50's% rate with a full-game handicapping system. Part of me isn't 100% joking either, I mean, we're getting out there with the standard deviations on the null hypothesis that these 2Hs are coinflips.

Anyway, equity was great again. All three sides had 1.5 points of equity (each is ~56% to win) the two totals had about 1 and 3.5 points equity, but of course <.500 once more. Nothing in late game went right again, each side blew a late lead, and two teams allow offensive rebounds with seconds remaining on FTs that prevent them from even trying to send it to OT. Std.

The much smaller late bets did well, 4-1, but I didn't have enough on them. I'm a little surprised at the equity I got betting that late - which I guess makes sense for lines that don't even come out until morning. I'll probably continue adding normal sized bets (mostly totals likely) in the mid-morning going forward even if I'm late to open by a couple hours. I probably won't get the best number, but I don't think being an hour late or so is akin to dart throwing anymore, after having looked back through all the movement over the last couple days for the late lines.

Akron $300 to win $2400 on Greek from a while back is worth ~$1450 right now, if you need to hedge:

MAC odds to win

Ball St.    .072

Buffalo    .379

B. Green  .172

Akron      .376

March Madness Bracket Odds

Lunardi's bracket is updating every day now. His new one is over on the right. Big winners recently (in Pom) are Zaga and WV.

Retrospective

Something I haven't really covered is the progression of equity throughout the year.

I'm not an authority on CBB, but I have pretty good knowledge derived from data analysis, plus I know a few people who have been at this longer than I have and do this professionally. And no, I'm not talking about random 50 year old guy who lives in Vegas, reads box scores all day, and makes decisions overusing concepts like "motivation" or "they're due", or rely on sample sizes of three games to make definitive conclusions. I mean the career lineshoppers with large valuable databases.

Very succinctly, CBB is pretty beatable, and it stays that way throughout the year (for totals at least). Totals are far easier to beat. Anyway, there's a reason limits are lower for CBB, and lower for totals. The first few months, through January or so, it's not uncommon to beat totals by 7+ points or sides by 2+ points on a fairly regular basis. In recent history, close isn't even efficient during this period. It's actually profitable to bet large lines moves AFTER the steam is finished, for many subsets. Lines just don't get moved enough. I don't know if that's a function of low limits and limited volume to sufficiently move lines, or what. In the 2nd half of the season, openers sharpen, 1 point on sides and 3 or 4 points on totals are more frequently the best movements of your day. Betting after steam is still >50%, but doesn't beat vig. Essentially, for a winning bettor, your equity against close for the first half of the year is understated, but it's a good barometer for the 2nd half of the season. This season bears out the dramatic change in the market pretty strikingly.

It hovers around 57%, and then crashes down to 54%. Last year was identical. My totals system has never changed. I've only paid more attention to recency biases after about a month at 53%, and gained about 1% of equity for it. I'm not sure what happens, like if some switch gets flipped at the line makers book and they go from 'thumb in the wind' or 'just use last years lines' to a more technical analysis with current season stats. Limits are low, so it's not terribly costly to let bettors sharpen your lines when data is scarce in the beginning of the year. Midway through, it's pretty easy to make a line. In the long run, over the 2nd half, my actual winning percentage shouldn't deviate much from the 54%, assuming I can keep getting the equity. 

I ran over expected (60%+) at the start, which happened last year as well, and has for many other people. Close just isn't efficient. Around the drop, I've typically run almost exactly at expectation, plus a very small handicapping edge from a close that still isn't truly efficient, but not exploitable on it's own either. I'm running a little below expectation over the last 75 or so because I can't buy a 1 or 2pt win and run really bad at avoiding or hitting overtimes. Going forward, 54-55% is roughly expectation.

The sides are a similar story. They get more action and have higher limits, making it progressively harder after the midseason magical switch flip. Unless you play just a game or two per day, or less, you likely don't get enough equity against close to cover vig. The value there is totally derived from a (hopefully) small handicapping edge that gets you around 53%. At this point in the year, for sides I'd really do better to just focus on lineshopping and get in before steam fully moves something. The expectation should be about the same if you're any good at it, and it's less work. But, I'm addicted to this stuff and don't have time to squat on Pinnacle and race to insta-hit any line that twitches.

Assuming Cal under loses (it's a huge dog right now): 1-3-1 today, with an OT under kill and flurry of points in final 13 seconds to push maryland.

Every total had 2-4 pts of equity, some of the biggest moves off open all day. And again, 1-4 vs. the 2H line that helps the game line obv, with the 1 win vs the 2H line already a huge favorite at the half. This has been going on for weeks, so frustrating.

Just for fun (massochism): Record on totals, last five weeks, final score 2.5pts or less from my lines: 4-14. For within 1.5 pts of my line, 1-8. Record was only even for those before the last five weeks. Big negative OT unders vs OT overs disparity too.

3/13

Wisconsin -1.5

Texas -3

BYU -1.5

Totals:

Louisville u143

LSU u140.5

Late morning adds (Getting in after movements, unlikey to get equity. Update: wow, these did get equity, I'm not used to this kind of movement past noon. Probably standard when games are added in the morning. System probably is still valid missing morning openers - I'll continue to bet and track late adds assuming I get up early enough to not totally miss the boat.)

Tenn u150

CSUN -2.5

LA Tech +10

ASU u140.5

Holy Cross +5

I'll prob be adding some sides in a bit. Pretty good equity on the totals, so I don't feel like a schmuck for going under heavy. Starting out with a Miami FL under win, and the Ball St. under just got blown up by an overtime. Running just insanely terrible in the last minute of these conference tournament games, if you need someone to miss/hit a FT, it's not happening.

Also FYI, if you got the bet, I'm hedging some of the Buffalo/Kent St. Game. That $300 to win $3600 is currently worth ~$925. Just betting enough to win $300 and recoup losses. If you lose you get most of the equity back in the implied worth of the Kent St. position. Either team will be big favorites (~8) against Ball St., then the final game will be another near coinflip against Miami OH or Akron in all probability. In that scenario it's worth more around $750-$800, but there's value in the chance Miami OH/Akron loses to Bowling Green or Ohio (inferior by several points).

I have a slightly +EV Miami OH position, so I'm not doing the same with Akron, which has less equity in it anyway (the $300 is worth ~$550).

3/12

VT u138.5

Maryland u143

Ball St. u116.5

Utah St. u134

California u139.5

Yeah - all unders, but totals openers to date for tournament games have generally been exhibiting about 3-4 pts+ under what I'd expect to see during conference play. Todays lines, for some of the tournys, aren't following the trend. I don't know what the deal is, but there's definitely been statistical justification for the several point drop in totals over prior years tournaments. Shrug, be wary of these possibly. 

3-1 totals today, good equity. Just a smidge of equity on sides, which went 1-2-1. Again, totals clearly have higher edges, I'm generally betting half units on sides and full units on totals. Come opening round after selection Sunday, that's definitely going to change. Spread lines get made for Joe Public too, and some start way off.

Conference Tournament Maket Odds

I'm using current Pinnacle (or BM for Thursday) lines for upcoming games plus the past two conference games lines as predictors of projected game lines for every possible matchup in the bracket, then dropping it in the tournament matrix. I'll keep tracking MAC for any of you who managed to get Kent St. +1200 or Akron +800.

The schedules and seeding are here: Conference Tournament Schedules

Google Documents Fair Odds for Tournaments:

Big East - Louisville easiest double bye path

SEC - 1pt HCA for Florida (games in Tampa)

Big 12 - 1pt HCA for Oklahoma and OK St. (games in OKC)

ACC

Big 10 - 1pt HCA for Indiana and Purdue (games in Indy)

Pac 10 - 2pts HCA for UCLA and USC (games in LA)

MAC

March Madness Bracket Odds

Most sites seem to be on hiatus with March futures right now, but Bracketology 101's new bracket will be up on the right in the next few minutes.

Update: Or not, they won't bother to put up the bracket, they just list teams and seeds and I don't feel like snaking through it to put everyone in the right region. They have some teams still in that seem ridiculous too, so I'm losing my faith in them. If anyone knows of a better prognasticator, let me know and I'll start using them. They make Lunardi look brilliant, and he doesn't even forecast remaining schedules.

Anyway, I am about to create some Google docs with market odds (well as best as I can do to predict future lines) on the conference tournments that have yet to start. I'll keep a running one for MAC too, in case you got in on Akron or Kent St. Hopefully it'll help spot some value, as some pretty square books have started to put up conference tournament futures.

Ugh, lots of 2H ridiculousness today. Running ~43% over the last few weeks on the 2H Pinnacle line that's the same side as the game bet. That's really hard to do. If I was a little more results oriented I'd switch to 1Hs, but I still have a few brain cells that aren't on tilt right now.

3/10

WKY -4.5

Seton Hall u131.5

3/11

Providence -9

UCF -3

Col St. -2

CSUF -2.5 (late add, just noticed UCRs injuries)

Providence o143.5

Texas u130.5

UCF o132

Marquette u141

Neutral court backwork done yesterday and just finished fatigue backwork, so I'm betting normal units on the totals.

FYI SEC tourny Odds changed to reflect yesterday's games and 1.5 pts HCA for Florida.

Also, I used the moneylines for tomorrow's MAC games for the first round games, then reconciled high level efficency stats to replicate the close of the last two conference MAC games as best I could. So, I'm now using those "adjusted adjusted" efficiency stats to create hypo game lines for all possible MAC games, and converting to a win%. The new odds are reflected in the old MAC post, and it's not really true KenPom anymore. I just wanted a better idea of my market equity on Akron and Kent St. should I need to hedge if lucky enough to get either into the finals.

3/10

Would bet Kent St. -11, if I hadn't maxed +1200 to win tourny.

Totals (teams didn't play yesterday)

Kent St. o132.5

Ohio o123.5

Weber St. o131.5

These include observed neutral court and tournament effects on totals. I'm making a scaling adjustment for that, but otherwise using the same totals system.

Bracket Odds / Lunardi

March Madness bracket odds for Lunardi's Monday release is over on the right. Zaga gets a huge boost overperforming ATS last night by 25pts.

3/9

I'm already invested in Zaga's win tonight, but Zaga -6.5 would be a play.

WKY -3.5

CoC -2.5

Niagara +4.5

Would also all be plays, but I've got fatigue backwork to do on disparate fatigue (bottom 3), so I'm not counting/betting these for now. I should be done with that sometime today.

Big 10 Tournament / KenPom

Illini  +480

Mich  +3900

M. St.  +250

Minn  +3300

NW  +4800

OSU  +1425

PSU  +5025

Purd  +290

Wisc  +780

Field  +7800

No value vs. Greek. The games are in Indianapolis. I have not included any HCA. If you assume Purdue has 2pts HCA, they're +190 fair, and then +250 on Greek has value. I don't know how much HCA the market will assign to these games.