What betting algorithm/formula do you use for predicting MLB games?

**jeffthehat1** · 06-10-14, 06:04 PM

Originally posted by Solid

I am trying to have an objective approach to betting, but I am curious of the thought process of others?

I would also like an honest response about how successful your bets are.

Also how many of you use computer programs to help predict a winner?

Dont need computer programs. Use a solid chase system and think out side the box. The reason 99% of bettors lose besides bad money mgmt. is the failure to think outside the box! But what do I know!

Good luck

**Solid** · 06-10-14, 07:11 PM

^ I don't chase games/series, I made a program that makes a prediction and checks the odds to see if it is worth it.

I just seem to be missing some x-factors (although there are many in baseball).

**SelanneMN** · 06-24-14, 10:15 AM

I actually just put together a formula late last week and tried it out from Saturday~Monday (40 games). Went 25~15 (63%).

The formula uses each team's winning %, current streak, last 10 and record differential for the home team and the away team.
(I put a point system together for the above. It's simple to do and I can explain it to you if you are ever interested).

In addition, I implemented a point system for each starting pitcher, by using the average ranking over the past 30 days from Yahoo fantasy sports. I just needed to get something in for the starting pitchers, and thought why not use what Yahoo provides based on their formula or player rankings.

Anyways, here are my picks for Tuesday using my formula:

STL, DET, BAL, SEA, TOR, MIA, OAK, CLE, LAD, PIT, MIN, SF, CIN, ATL, MIL

In all honesty, I probably would have picked most of the above teams without using the formula, except for MIN, my home team.

I'll continue to share my picks so you can see how it goes.

FYI...before I did this formula, I usually bet just moneylines on first place teams when their top 1 or 2 pitchers were on the mound.
I usually bet just 1~3 games a day.

**RollinDo** · 06-24-14, 10:23 AM

There's this thing called a coin...you flip it...usually works better than trying to use logic and what makes sense.

**posey** · 06-24-14, 10:35 AM

@Selanne:
If you don't mind I would be interested in how you set it up. I made a few tries to set up something similar but failed after some days really hard.

**flying lotus** · 06-24-14, 02:15 PM

@Selanne I would definitely be interested in hearing about your system/equation - I've been working on some things as well, but nothing has come out to be successful enough for me to start putting actual wagers down

**akphidelt** · 06-24-14, 02:27 PM

Come join the collaborative system thread I've started. Combining all these formulas and ideas in to one giant system would be awesome.

**bigben1985** · 06-24-14, 03:46 PM

Originally posted by jeffthehat1

Dont need computer programs. Use a solid chase system and think out side the box. The reason 99% of bettors lose besides bad money mgmt. is the failure to think outside the box! But what do I know!

Good luck

I think you know exactly what you are talking about. The chase systems that are most successful can take about 10-30 minutes of prep work per day. Then 8 hours of waiting for the games to start.

**SelanneMN** · 06-24-14, 10:02 PM

I would be glad to share it with you. It's easy to do, but a little difficult to explain. I hope this will make sense and won't be too confusing.

Part 1.

What I use is Excel, Espn.com and Yahoo Fantasy Baseball.

Create a spreadsheet like below. Put the away team on top and the home team on the bottom.
(FYI...the following examples were used for last Saturday's games).

Team W% Streak Last 10 Away/Home S.Pitcher Total
SF 589 -60 10 50 1153 1742
ARZ 416 10 30 -130 357 683

Step 1 Using ESPN.com for the next 4 steps, enter the team winning percentage for each team, omiting the decimal.
Ex: SF winning percentage was .589 Enter 589 for SF (no decimal) and ARZ 416.

Step 2 Add 10 points for each game on a winning streak and -10 pts for each game on a losing streak.
Ex: SF was on a 6 game losing streak. Enter -60 for SF and 10 for ARZ with a 1 game winning streak.

Step 3 Add 10 points for each win in their last 10 games.
Ex: SF was 1-9 so enter 10 points for SF. ARZ was 3-7 in their last 10. Enter 30 for ARZ.

Step 4 For Away and Home, add 10 points for each game above .500 and -10 pts for each game below .500.
Be sure to check the AWAY record for SF and the HOME record for ARZ.
Ex: SF was 5 games above .500 on the road, enter 50 points. ARZ was 13 games below .500 at home, enter -130.

Step 5 For this step (S.Pitchers) I use my free yahoo fantasy baseball account and go to "Players" or "Player List"
From here you will need to sort the players as follows... Under Status, select "All Players." Under Position, select "Probables."
Under Stats, select "Last 30 days avg." Next, click Filter. Finally, sort the players by clicking on "Rank" (Not O-RANK!!!).
Find each starting pitchers ranking and put that number in the opposing teams box for S.Pitchers. Yahoo ranks the best players
with lower numbers. For example, Kershaw was recently ranked 1 this week. Therefore, you want to give Kansas City 1 point and
give the Dodgers whatever the KC pitcher's rating was.

Once all the above data has been entered for each category, add up the points. The team with the most points is the favored
team. This is what I used for the games from last Saturday~Monday and actually was 24~16 (not 25~15 as originally mentioned.
I put the Giants as winners over Arizona by mistake). Saturday was 9&6, Sunday 9&6 and Monday 6&4. Only 60%, so nothing too
special. There are a few more steps to the formula which I will explain later, along with something I discovered when playing
Monday morning QB. The next few steps will help narrow down which games to focus on if only betting on a few games.

Will post more later...

**SelanneMN** · 06-25-14, 08:27 AM

Here is Part 2. (See above for Part 1)

Step 6) Create a team point list. Using the total points you tallied for each team from steps 1~5, simply rank them in order from the team with the most points, to the team with the fewest points, regardless of which team is favored or the underdog. Just simply rank every team from highest to lowest. The top 5 teams will be of importance in the next step or two.

Yesterday's games (6/24) were as follows...
DET 2130
STL 2026
SEA 1612
ATL 1544
HOU 1494

Wednesday's games (6/25)
SEA 2094
STL 2088
WAS 1999
DET 1798
LAD 1746

Step 7) Create a team match up point differential list. Using the total points from steps 1~5, subtract the underdogs total points from the favored teams total points for each match up. Then, create a match up differential list by putting the team match ups with the biggest point differential first, second most second, and so on. The top 3 match ups will be of importance in the next and final step.

The following were from Tuesday's games.

1110 STL VS COL (STL was favored)
863 DET VS TEX (DET was favored)
827 CWS VS BAL (BAL was favored)

Here are for Wednesday (6/25) games...
1379 DET VS TEX (DET is favored)
1325 SEA VS BOS (SEA is favored)
1232 WAS VS MIL (WAS is favored)

Step 8) If any of the top 3 favored teams in the team match up list (Step 7) are also listed in the top 5 team point list (Step 6), then I would bet on them. Looking over Tuesday's results first, DET and STL were in both lists, however, BAL was not (they were 6th and not listed. They actually lost anyways). I placed a bet on both DET and STL yesterday and went 1 for 2.

Looking at the results for today's games, DET, SEA and WAS are in both lists. I placed my moneyline bets on all 3.

After playing Monday QB so to speak, I would have been 7 wins and 3 losses had I had used this strategy to zoom in on better teams with more favorable match ups.

My results since I started last Saturday would have been this...

Saturday: MIN, SF and TB were all 3 winners. (3~0).
Sunday: LAD win but NYY lose (Tanaka vs BAL). (1~1)
Monday: STL & MIA win, but MIL loses at home to WAS (2~1)
Tuesday: DET wins and STL loses in COL. (1~1)

Total of 7~3 for 70%. Still way too early and too small of a sample size to tell if this will continue. For the next several days I'll continue to post my picks before the games, so you can see what I am betting on and if this formula/strategy actually has any value.

Picks for Wednesday on all games: (Steps 1~5)
DET, SEA, WAS, STL, NYM, LAD, TOR, CWS, MIA, LAA, SD, ARZ, CIN, PIT & ATL

DET, SEA, WAS (Steps 1~8)

Somethings to note. I don't have any weight on any categories, nor did I include ERA or Avg Runs Per Game, but I might look into later. I want to sample more games this way first.

Also, if a pitcher has no rating from Yahoo, such as the SD Padres rookie that started Monday and beat the Giants, I gave him a rating of 1500, or actually put that in SF Giants S.Pitching box.

Finally, on days when there are only a couple of games scheduled, I would either not place a bet or only do so when I see that a team has a huge match up point differential (> 800 pts) and their total points is (>1500).

My picks today (just using Steps 1~5 to determine which teams were the favorite & underdog) were terrible. I went 6~9 for a running total now of 30 & 25 for 54%. It was the first in 4 days of being less than 50% on the day. My goal is to stick to 1~3 bets maximum a day using the 8 Step process and staying above 70%.

Hope this wasn't too confusing and if you have questions please let me know.

Best of luck !!

**posey** · 06-25-14, 01:32 PM

Thx 4 your explanation.

**sportsfreak24** · 03-30-19, 07:17 PM

Hello,
I am sorry to be bothering you. I came across a post you left back in 2014 involving a formula that helped you predict baseball scores. You said you went 25-15 and were willing to explain it someone that reached out? Is there anyway you can find some time to explain that to me? I am just getting into sports betting and any help would be greatly appreciated. You seem to have a lot of knowledge with it and I could not be anymore thankful if you found the time to explain the formula to me. I am never on forums like this but when i came across this i couldnt pass it up! Thank you so much!

**gojetsgomoxies** · 03-31-19, 11:58 AM

i'm new to baseball betting but a fanatic fan at many times in past.

anyway, i would just model the starting pitcher myself....... as dan shan said in another thread, let the market determine a decent line and then work at the periphery of it (analyzing starting pitchers)

i am looking for new good pitchers, good pitchers on bad teams and vice-versa

**bettingman6** · 03-31-19, 01:54 PM

Fangraphs gives you some nice info that can tell you what teams and pitchers are overrated or underrated.

It's usually better to take underdogs earlier in the year. There's some evidence that teams that were supposed to be bad in the preseason do decently in April, but fade starting in May.

**gojetsgomoxies** · 04-01-19, 12:33 PM

Originally posted by bettingman6

Fangraphs gives you some nice info that can tell you what teams and pitchers are overrated or underrated.

It's usually better to take underdogs earlier in the year. There's some evidence that teams that were supposed to be bad in the preseason do decently in April, but fade starting in May.

that makes sense on april, and the subsequent fade. lots of youthful energy in april, and then beaten up by so many losses by july/august.

the fangraph sounds interesting. someone mentioned greinke losing velocity this season. i think things like that are like free money.