moncada and choi 1.5 -114 now ez game

The Bauer thing depends if you buy into the BetBuilder Efficient Market theory

Right now Boyd is u6 -190
Lopez o5.5 -109
Bauer u7.5 -114
Glasnow o6.5 -130

all in line with my projections. If it does become efficient we good. If not we screwed with Bauer.

Efficiency within Efficiency lemma to the Game within the Game theorem.

I had Albies u2.5 yesterday, closed at 1.5, lost. Seemed impossible.

One to glance at is Rodriguez. I have u4.5 +102 as fair. Sitting at +100 now. Hoping for some steam chasers on the over. Is Taiwan awake yet?

This is why I'm worried tho.

One HRR H2H offered by Jazz: Moncada -110 over Adames -120.

Adames is batting 9th and I have him projected at 75% of Moncada (who is batting 2nd). Yet Adames is the favorite at -120. Just the fact that Moncada should get an extra AB should be enough to make him the favorite. Also why are they matching up #2 batter vs #9

Same with Myers -110 over Posey -120 and Kinsler -110 over Crawford. Not as big of favorites as Moncada, but still should be -140 or so.

Rain a concern in the TB game as well. Not sure how to account for that in the theorem

Game within the game. Rays a favorite therefore all players on rays regardless of talent or batting order shall be a favorite over the other teams players

Here are two more new TB plays.

Aaron "My name is" Judge -105 over Springer (the we won yesterday with Springer over Judge, so let's reverse it because he can't do it twice in a row theory)

K. Davis -115 over Trey Mancini (the cover play so we don't get banned since Mancini is a lock to always win theory)

3-1-1 yesterday, +2.05
YTD 7-10-2 -4.09

how much total storage do you have DNRS? daily archives and open/closing lines must add up, you got racks in the basement or something?

S CASTRO (MIA) TOTAL BASES +105 (S CASTRO (MIA) TOTAL BASES vrs J VOTTO (CIN) TOTAL BASES) +105
A RENDON (WAS) TOTAL BASES -105 (A RENDON (WAS) TOTAL BASES vrs R HOSKINS (PHI) TOTAL BASES) -105
T MANCINI (BAL) TOTAL BASES -115 (K DAVIS (OAK) TOTAL BASES vrs T MANCINI (BAL) TOTAL BASES) -115
M KEPLER (MIN) TOTAL BASES EV (M KEPLER (MIN) TOTAL BASES vrs M CONFORTO (NYM) TOTAL BASES) EV

holding my nose a bit on castro and mancini but gotta go with it. you and me are opposites on mancini, DNRS. for mancini, i think i know why we differ, i'm likely docking HRs too much. I figure him for more TBs on average, but dingers are an outsize contributor there.

Not that much. My total MLB Lines SQL file is only ~2700 megs and it has every line move from Pinnacle, 5D, and bookmaker for every bet offered since 2004

My setup is all VPS, mostly through Linode.

For lines: 4gb ram/80 gb SSD

For modeling baseball/basketball/other sports, storing my bets, grading: 8gb ram/160 storage (the lines DB is also copied here so I don't need to keep querying it - and there doesn't need to be checks if the line already exits as it's only inputting it if it meets the criteria on the lines DB which is a little more resource intensive). with all the projections archived it's currently using 101 gigs. I'm thinking about archiving pre 2016 ones to a different site as i've noticed slow downs in the machine learning modeling.

For harness racing: 4gb ram/80gb SSD for data, another one for machine learning

For thoroughbred racing: Same as harness

Then I have one smaller 2 gb VPS that serves all the front ends for these.

There are a couple clones at various places in case something goes down (digital ocean and BuyVM)

It costs only $130/month for all of them and backup services and is a business write off.

The biggest cost is horse racing data. It costs $1060/month but I get real time pool data from all tracks except the Canada ones are weird. They will give you total pool numbers but not real time odds or even closing odds. It's only an issue with Woodbine as that's the only Canada track that is worth betting.

I pay $0 for the data for lines / major sports data. Just scrape it all.

TB/CWS Meadows -120 over Abreu
TB/CWS Pham -120 over Jimenez

sounds slick. what kind of rebates are you getting on the racing? isn't woodbine the track that american ADWs wire through to get asian races?

Woodruff u5 -114 (Fair: -305 so bad)
McHugh o7 +120 (fair +104)
Strailey o4 -114 (fair -165)
Pena u5.5 -114 (fair -149)
Flaherty u6 -114 (fair -203)
Paxton u6.5 -114 (fair -189)
Hellickson u4 -114 (fair -219)
Lyles u4.5 -114 (fair -143)
Lynn u5 -114 (fair -236)
PIvetta u6 -114 (fair -273)
T. Richards u5 -114 (fair -236)
Yu Darvish u6 -114 (fair -378)

Some really bad lines out there IMO. Curious if it matches up to what others have. Lot of unders. I think it's an over reaction to increased K's narrative.

It depends on the track. Some of the Penn ones I can get 17% on exotics and 10% on WPS, some of the NYRA ones it's basically 0% (I bet those in the sportsbooks in Vegas for those sweet comp dollars and freeplays). The Ohio ones are good for rebates too, especially harness.

What I'm doing is sorta illegal, but it doesn't matter because my SN isn't attached to my real name. I have a condo in New Mexico, New Mexico citizenship and use a ADW. In the house there's just computers and I set up an API that I just feed bets through and it pushes it to the ADW.

It looks weird because there's literally 0 furniture in the place, just computers. I'm sure the cops would think it's a terrorist thing. I've been there once in the last 3 years.

it's cool to see someone winning at racing that's not on an industrial scale, that is no small feat. after being tossed from so many places for use of a vpn or bot or whatever, i've always toyed with the idea of setting up a laptop in a condo like that and mounting a mechanical overlay on top of the keyboard with little mechanized pokers for each key, complete with a little moving stylus for the trackpad. zero software solution to get around geographic restrictions.

Most ADWs now have API's to accept bets, you can just set up a server to push them through it or you can remote desktop in.

Amwager is working towards incorporating pool data into theirs which would be great, but horse racing data is so monopolized by trackmaster I doubt they'll be able to do it for free.

I think the new big "whales" of corporations introduced weird dynamics into the game. It used to be beneficial to bet last second, but I think betting 5 minutes before post is better now. You see this in China too. There's a lot more 1/9, 1/5 especially in harness, but sometimes they go overboard and create a lot of inefficiencies.

Been hitting a lot of mid odds like 7/1 - 12/1 because the 1/9 horse breaks or just doesn't show up and some big corp dumped way too much money on it.

Best I saw was there was I think maiden stakes race at The Red Mile and someone dropped about 50k on this one horse to show. The Red Mile doesn't even get 50k on show combined across all races in a typical race day. The horse wasn't that good, it was a 10 horse race, and I remember the horses were all really young and didn't have a lot of experience. Dropped $100 to show and place on every other horse, the 50k show horse breaks coming out of the gate.

Top 3 paid

84.40/110/38.20
0/177.6/150
0/0/90.6

Free 27k

https://www.trackinfo.com/results.js...HLEX08/16/2018

keep looking, there's a much much better line for mchugh out there

when i was more interested in it, i'd play around in the smaller show pools, especially when they'd go negative and everyone would go berserk to get their 2.10 or 2.20 minimum parimutuel payout. not enough hours in the day for me. i am too disorganized.

I'd watch some of the smaller harness tracks like Saratoga, The Red Mile, etc. I've seen a lot of bridge jumping there, especially in stakes races. Thoroughbred you see a lot of ~$8-$14 winners, at these tracks it's either $2.20 or $109

Maybe we were on right track with H2H HRR since Kinsler is now -155 and Myers -135, tho Moncada is still -110. 2 out of 3 ain't bad tho

Also got crushed on the McHugh line, down to 6.5. Lets hope society is wrong

We should start arb'ing Jose Ramirez TB bets with "Will Jose Ramirez get a hit" on the NO.

Four bets, 0 hits.

Waiting for a nikolas bloop now


Sent from my iPhone using Tapatalk

local total base lines

https://pastebin.com/43fW6Muz

The mound at the White Sox game is all broken and the pitchers keep slipping, how do we price that into our models

Thanks man, I was inspired to head back into the lab today for some HRR R&D so I'm just getting to this thread.

K. Davis -110 over Mancini
Yelich -136 over Pujols

Risking 2.46u

Squad, do any of these HRR seem ridiculous to you?

Merrifield -110 over Santana -120
Mondesi -110 over Healy -120
Andrus +100 over Ahmed
Rendon +105 over Realmuto

Any database developers have easy access to a csv file of historical player box scores from the last year or so? something like this

date, team, player, pa, ab, r, h, rbi, hr, sb, k

want to put together a presentation for our sympolsium to flesh out the hrr distribution some more to convert between mean projection and median line and value half points correctly

Santana NA, Merrifield 2.55
Mondesi 2.15, Healy 2.18
Andrus 2.13, Ahmed 1.58
Rendon 2.64, Realmuto 2.30

I can write a script tomorrow to export that. Do you just want an insanely long list that would be for every player? So like 162 games per player?

Let's buy back some of this steam other people doing

Mahle u5.5 +110 (fair -112)
Syndergaard u6.5 -108 (fair: -144)
Maeda o6 +110 (fair -124)

Very bittersweet Willy Adames hits a single top 9 to make the game go over. I will never doubt him again.