Is the relationship between field size and profits non-linear?

[lansdale]

Virtually everyone I know who plays this game would prefer larger fields, and I know that people like Jeff P. and Ken Massa have written about the profitable aspects of larger fields, but I was curious as to whether the difference in profit between, say, an 8-horse and a 10 or 12-horse field is as dramatic as it seems to me. So, although I'm anything but a mathematician or a prob. & stat. guy, I wondered whether there was any research out there indicating nonlinearity. I really am just looking to back up my instinct that taking even a second look at most small fields is anything but a waste of time. Any help appreciated.

Cheers,

lansdale

[Jeff P]

Field size all by itself will not lead to profits. But as a general rule larger fields tend to help a player's roi.

Here's a data sample spanning calendar years 2006, 2007, and 2008. The sample is limited to only the top ranked horse for JCapper JPR in each race.
UDM Definition: JustJPR
Divisor: # UDM Def Divisor: 999
Surface Req: *ANY Surface*
Distance Req: *ANY Distance*

JPR: MinRank= 1 MaxRank= 1
MinVal= -999 MaxVal= 999
MinGap= -999 MaxGap= 999
Running Style: ALL

Data Window Settings:
Divisor = 999 Odds Cap: None
Filters Applied:

Surface: (ALL*) Distance: (All*)
From Index File: C:\2008\pl_678_JPR1_Master.txt


Data Summary Win Place Show
Mutuel Totals 132365.70 133226.10 130886.00
Bet -146672.00-146672.00-146672.00
Gain -14306.30 -13445.90 -15786.00

Wins 21587 35993 45269
Plays 73336 73336 73336
PCT .2944 .4908 .6173

ROI 0.9025 0.9083 0.8924
Avg Mut 6.13 3.70 2.89

Here's how the sample looks when the data is further broken out by field size:
By: Field Size

Field Size Gain Bet Roi Wins Plays Pct Impact
1 0.00 0.00 0.0000 0 0 .0000 0.0000
2 -8.00 8.00 0.0000 0 4 .0000 0.0000
3 -40.70 158.00 0.7424 36 79 .4557 1.5481
4 -267.80 1712.00 0.8436 367 856 .4287 1.4565
5 -1013.30 8546.00 0.8814 1630 4273 .3815 1.2959
6 -2451.70 21528.00 0.8861 3635 10764 .3377 1.1472
7 -2472.00 28056.00 0.9119 4352 14028 .3102 1.0539
8 -2922.60 26998.00 0.8917 3859 13499 .2859 0.9712
9 -1684.30 21182.00 0.9205 2931 10591 .2767 0.9402
10 -1895.40 18910.00 0.8998 2418 9455 .2557 0.8688
11 -691.80 9148.00 0.9244 1125 4574 .2460 0.8356
12 -878.20 8664.00 0.8986 1022 4332 .2359 0.8015
13 -31.30 906.00 0.9655 115 453 .2539 0.8624
14 52.00 848.00 1.0613 96 424 .2264 0.7692
15 0.00 0.00 0.0000 0 0 .0000 0.0000
16 -2.00 2.00 0.0000 0 1 .0000 0.0000
17 0.00 0.00 0.0000 0 0 .0000 0.0000
18 0.00 0.00 0.0000 0 0 .0000 0.0000
19+ 0.80 6.00 1.1333 1 3 .3333 1.1324
If, from the above sample you get the idea that field size is a factor that can sometimes be used synergistically with other factors to help move you towards profitability - then I've managed to get the point across.


-jp

.

[HUSKER55]

I am at the point of not betting into any field that does not have 8 horses. The favorite and exoctics are bet into oblivion and last place is not that great either

[lansdale]

Field size all by itself will not lead to profits. But as a general rule larger fields tend to help a player's roi.

Here's a data sample spanning calendar years 2006, 2007, and 2008. The sample is limited to only the top ranked horse for JCapper JPR in each race.
UDM Definition: JustJPR
Divisor: # UDM Def Divisor: 999
Surface Req: *ANY Surface*
Distance Req: *ANY Distance*

JPR: MinRank= 1 MaxRank= 1
MinVal= -999 MaxVal= 999
MinGap= -999 MaxGap= 999
Running Style: ALL

Data Window Settings:
Divisor = 999 Odds Cap: None
Filters Applied:

Surface: (ALL*) Distance: (All*)
From Index File: C:\2008\pl_678_JPR1_Master.txt


Data Summary Win Place Show
Mutuel Totals 132365.70 133226.10 130886.00
Bet -146672.00-146672.00-146672.00
Gain -14306.30 -13445.90 -15786.00

Wins 21587 35993 45269
Plays 73336 73336 73336
PCT .2944 .4908 .6173

ROI 0.9025 0.9083 0.8924
Avg Mut 6.13 3.70 2.89

Here's how the sample looks when the data is further broken out by field size:
By: Field Size

Field Size Gain Bet Roi Wins Plays Pct Impact
1 0.00 0.00 0.0000 0 0 .0000 0.0000
2 -8.00 8.00 0.0000 0 4 .0000 0.0000
3 -40.70 158.00 0.7424 36 79 .4557 1.5481
4 -267.80 1712.00 0.8436 367 856 .4287 1.4565
5 -1013.30 8546.00 0.8814 1630 4273 .3815 1.2959
6 -2451.70 21528.00 0.8861 3635 10764 .3377 1.1472
7 -2472.00 28056.00 0.9119 4352 14028 .3102 1.0539
8 -2922.60 26998.00 0.8917 3859 13499 .2859 0.9712
9 -1684.30 21182.00 0.9205 2931 10591 .2767 0.9402
10 -1895.40 18910.00 0.8998 2418 9455 .2557 0.8688
11 -691.80 9148.00 0.9244 1125 4574 .2460 0.8356
12 -878.20 8664.00 0.8986 1022 4332 .2359 0.8015
13 -31.30 906.00 0.9655 115 453 .2539 0.8624
14 52.00 848.00 1.0613 96 424 .2264 0.7692
15 0.00 0.00 0.0000 0 0 .0000 0.0000
16 -2.00 2.00 0.0000 0 1 .0000 0.0000
17 0.00 0.00 0.0000 0 0 .0000 0.0000
18 0.00 0.00 0.0000 0 0 .0000 0.0000
19+ 0.80 6.00 1.1333 1 3 .3333 1.1324
If, from the above sample you get the idea that field size is a factor that can sometimes be used synergistically with other factors to help move you towards profitability - then I've managed to get the point across.


-jp

.

Hi Jeff,

Much thanks for the great charts. Have to say I'm surprised that there's only an incremental difference in ROI above seven horses, but this is exactly what I was looking for. It could be that the large difference in ROI I find in larger fields is simply a function of my handicapping, which is more one-dimensional than the comprehensive approach encapsulated in the JPR figure. BTW, not to seem too crazy, but did you notice that odd-numbered field sizes above seven have somewhat better ROI those that are even-numbered?

Thanks again.

Cheers,

lansdale