Foundations of Database Handicapping

 

 

Benchmark Testing

My purpose in writing this Help Document is to help you understand how Data Window research can help you in your handicapping. Not all horses that enter a starting gate have an equal historical probability of winning the race. Some horses have advantages over today’s field. And others have disadvantages. Understanding these advantages and disadvantages, and using that understanding in your live play, is what database handicapping is all about.

 

I think looking at benchmark tests of JCapper factors helps new users to shorten the learning curve. If nothing else, benchmark tests serve to give both context and meaning to the factors found in the program. In this document I’m not going to examine every factor in the program – only some of them. But after you’ve built a few databases of your own using your Bris or TSN files you’ll certainly have the ability to use the Data Window to run your own tests on as many factors as you like.

 

As the author of a handicapping program like JCapper, I have found one caveat to be true about database handicapping. It is far easier to create profitable UDMs when using a good starting point (as opposed to using a bad one.) It should be obvious that good starting points in your handicapping relate to high win percentages and/or high roi. Horses with historically low win percentages and/or low roi usually make terrible starting points for UDMs.

 

After reading this document the value of database handicapping should start to become apparent to you.

 

The following benchmark tests were taken from actual Data Window queries of my own calendar year 2006 database. All data samples presented in this Help Doc are based on data found in Bris $1.00 Single Format DRF Data Files.

 

ALL Starters in the Database

I always think it’s a good idea to begin any discussion of historical thoroughbred racing data with a sample that includes all horses in the database. This gives you a point of reference when you begin your handicapping. If you are doing a good job in your handicapping, horses you select will substantially outperform horses selected at random. When you run a Data Window query using the ALL Button, the Data Window shows you a recap of all starters in the database. In JCapper, the ALL Button (all starters in the database) is the closest thing you have to horses selected at random. Using the ALL Button, here’s what my calendar year 2006 database looks like:

 

     Data Window Settings: (RUN 3/18/2007)

     999 Divisor

     Surface: (ALL*)  Distance: (All*)

     (From Index File: D:\2007\Q1_2007\pl_Complete_History_06.txt)

 

 

     Data Summary         Win     Place      Show

     Mutuel Totals  309882.70 305170.00 303211.00

     Bet           -404684.00-404684.00-404684.00

     Gain           -94801.30 -99514.00-101473.00

 

     Wins               24997     49798     73448

     Plays             202342    202342    202342

     PCT                .1235     .2461     .3630

 

     ROI               0.7657    0.7541    0.7493

     Avg Mut            12.40      6.13      4.13

 

 

 

Post Time Odds vs. JPRToteProb

 

Post Time Odds

It has been said that the betting public makes a pretty good opponent. No, I’m not talking about each individual bettor. I’m talking about the collective intelligence of all bettors everywhere. As each race is bet, tens of thousands of bettors each weigh and apply the dozen or so (in some cases more) handicapping factors they think will shape the outcome of the race at hand. Collectively, they reach, by my rough estimate, somewhere between one and two million decision points as they handicap each race. As long as there has been modern day thoroughbred racing, post time favorites have won approximately 33 percent of all races. However, in recent years, because average field size has been shrinking, the win percentage of post time favorites has gone up slightly higher. My calendar year 2006 database shows that post time favorites (with no attempt to break ties for post time favoritism) won almost 35 percent of all races. 81.47 percent of all races were won by one of the first four choices in the betting. That should tell you quite clearly that quite a bit of collective intelligence exists in the odds.

 

 

     By: Odds Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -8925.70  52064.00    0.8286   9103  26032   .3497     2.8306 

        2   -9449.50  49120.00    0.8076   5120  24560   .2085     1.6875 

        3   -9790.00  49288.00    0.8014   3705  24644   .1503     1.2170 

        4  -10366.90  49148.00    0.7891   2609  24574   .1062     0.8594 

        5  -10250.00  48728.00    0.7896   1865  24364   .0765     0.6196 

        6  -11820.00  45690.00    0.7413   1162  22845   .0509     0.4117 

        7  -10164.50  38204.00    0.7339    699  19102   .0366     0.2962 

        8   -8003.30  28648.00    0.7206    377  14324   .0263     0.2130 

        9   -4383.80  19734.00    0.7779    225   9867   .0228     0.1846 

       10   -6529.30  12770.00    0.4887     75   6385   .0117     0.0951 

       11   -2924.70   6686.00    0.5626     37   3343   .0111     0.0896 

       12   -1397.80   3636.00    0.6156     18   1818   .0099     0.0801 

       13    -564.00    640.00    0.1188      1    320   .0031     0.0253 

       14    -219.80    316.00    0.3044      1    158   .0063     0.0512 

       15      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       16      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       17      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       18      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       19      -4.00      4.00    0.0000      0      2   .0000     0.0000 

 

 

JPRToteProb

JPRToteProb is the result of sending JPR (JCapper Power Rating) and post time odds into an algorithm that calculates a probability after the odds are known. This algorithm combines the collective intelligence of the betting public with JPR and produces a very accurate probability.

 

How accurate is this probability?

 

My calendar year 2006 database shows that JPRToteProb is even more accurate than the probability inherent in the odds set by the betting public. The top three ranked JPRToteProb horses each won a higher percentage of their races than the top three horses ranked by post time odds. One of the top four JPRToteProb horses won 82.25 percent of all races in the database. Further, the top four ranked JPRToteProb horses had a higher flat bet win roi than the top four horses ranked by post time odds.

 

     By: JPRToteProb Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -7823.60  49682.00    0.8425   8755  24841   .3524     2.8529 

        2   -9145.30  49708.00    0.8160   5325  24854   .2143     1.7343 

        3   -9542.30  49636.00    0.8078   3735  24818   .1505     1.2182 

        4  -10626.60  49666.00    0.7860   2615  24833   .1053     0.8524 

        5  -10582.70  48998.00    0.7840   1877  24499   .0766     0.6202 

        6  -11837.80  45960.00    0.7424   1189  22980   .0517     0.4188 

        7   -9910.10  38426.00    0.7421    728  19213   .0379     0.3067 

        8   -6129.90  28704.00    0.7864    440  14352   .0307     0.2482 

        9   -7719.10  19816.00    0.6105    194   9908   .0196     0.1585 

       10   -5898.00  12806.00    0.5394     84   6403   .0131     0.1062 

       11   -3671.20   7076.00    0.4812     34   3538   .0096     0.0778 

       12   -1622.40   3388.00    0.5211     16   1694   .0094     0.0765 

       13    -156.50    586.00    0.7329      4    293   .0137     0.1105 

       14    -123.80    220.00    0.4373      1    110   .0091     0.0736 

       15      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       16      -4.00      4.00    0.0000      0      2   .0000     0.0000 

       17       0.00      0.00    0.0000      0      0   .0000     0.0000 

       18      -4.00      4.00    0.0000      0      2   .0000     0.0000 

       19      -2.00      2.00    0.0000      0      1   .0000     0.0000 

 

 

My calendar year 2006 database broken out by JPRToteProb numeric value is shown below. If nothing else this next chart should give you a clear sense of the algorithm’s accuracy. The probability range for each row is indicated by the values in the min and max columns. The actual win percentage along with number of wins, plays, and flat bet win roi is also shown. For example, the top row shows data for horses where the algorithm said the probability of winning the race was between 0 and 5 percent. The win rate achieved for the 62,157 horses in that row was actually 2.63 percent.

 

     By: JPRToteProb

 

     >=Min       <Max      Gain       Bet       Roi   Wins  Plays     Pct   Impact

   -999.00       0.05 -41396.00 124314.00    0.6670   1634  62157   .0263   0.2128

      0.05       0.10 -21179.40 100082.00    0.7884   3606  50041   .0721   0.5833

      0.10       0.15 -10819.40  54184.00    0.8003   3346  27092   .1235   0.9997

      0.15       0.20  -8354.80  45586.00    0.8167   3927  22793   .1723   1.3946

      0.20       0.25  -4676.10  26008.00    0.8202   2923  13004   .2248   1.8195

      0.25       0.30  -3284.80  18180.00    0.8193   2470   9090   .2717   2.1995

      0.30       0.35  -2538.80  15730.00    0.8386   2504   7865   .3184   2.5771

      0.35       0.40  -1054.10   8024.00    0.8686   1514   4012   .3774   3.0547

      0.40       0.45   -845.00   6616.00    0.8723   1460   3308   .4414   3.5726

      0.45       0.50   -471.80   4054.00    0.8836   1028   2027   .5072   4.1052

      0.50       0.55    -66.60    722.00    0.9078    208    361   .5762   4.6640

      0.55       0.60   -109.80   1104.00    0.9005    349    552   .6322   5.1178

      0.60       0.65     -4.80     78.00    0.9385     27     39   .6923   5.6040

      0.65       0.70      0.10      2.00    1.0500      1      1  1.0000   8.0947

      0.70       0.75      0.00      0.00    0.0000      0      0   .0000   0.0000

      0.75       0.80      0.00      0.00    0.0000      0      0   .0000   0.0000

      0.80       0.85      0.00      0.00    0.0000      0      0   .0000   0.0000

      0.85       0.90      0.00      0.00    0.0000      0      0   .0000   0.0000

      0.90  999999.00      0.00      0.00    0.0000      0      0   .0000   0.0000

 

 

 

 

Morning Line Odds vs. JPRMLProb

 

 

Morning Line Odds

The morning line odds are set by a track employee known as the morning line oddsmaker. The morning line is not necessarily designed as an attempt to pick the winner of each race. Instead, the job of the morning line oddsmaker is often twofold: First, the morning line can be a prediction of how the public will bet the race. And second, it can be an attempt by the morning line oddsmaker to maximize betting on the race for the track. Nonetheless, many morning line oddsmakers are bright talented handicappers. My calendar year 2006 database shows that there is quite a bit of inherent predictability in the morning line odds. Nationally, the morning line favorite won approximately 31 percent of all races.

 

     By: Morning Line Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1  -10705.00  51096.00    0.7905   7963  25548   .3117     2.5230 

        2   -9762.10  51474.00    0.8103   5140  25737   .1997     1.6166 

        3  -10119.20  54976.00    0.8159   4082  27488   .1485     1.2021 

        4  -13038.50  54920.00    0.7626   2933  27460   .1068     0.8646 

        5  -10115.10  50654.00    0.8003   2045  25327   .0807     0.6536 

        6  -11783.40  44698.00    0.7364   1247  22349   .0558     0.4517 

        7   -8563.30  37392.00    0.7710    828  18696   .0443     0.3585 

        8   -7812.10  26848.00    0.7090    435  13424   .0324     0.2623 

        9   -5815.80  16902.00    0.6559    203   8451   .0240     0.1944 

       10   -3507.60   9406.00    0.6271     84   4703   .0179     0.1446 

       11   -2728.60   4278.00    0.3622     25   2139   .0117     0.0946 

       12    -468.60   1658.00    0.7174     12    829   .0145     0.1172 

       13    -284.00    284.00    0.0000      0    142   .0000     0.0000 

       14     -86.00     86.00    0.0000      0     43   .0000     0.0000 

       15     -12.00     12.00    0.0000      0      6   .0000     0.0000 

       16       0.00      0.00    0.0000      0      0   .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       0.00      0.00    0.0000      0      0   .0000     0.0000 

 

 

 

JPRMLProb

JPRMLProb is the result of feeding JPR (JCapper Power Rating) and the Morning Line Odds into an algorithm that calculates a probability before the odds are known. This probability has proven itself to be accurate across large data samples as evidenced by the chart below. Note that the win percent parallels that of morning line odds rank – with one important difference: The flat bet win roi for the top four ranked JPRMLProb horses taken as a whole significantly outperforms the top four ranked Morning Line Odds horses.

 

     By: JPRMLProb Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -5510.00  49654.00    0.8890   7678  24827   .3093     2.5034 

        2   -6193.70  49664.00    0.8753   5000  24832   .2014     1.6299 

        3  -10333.50  49672.00    0.7920   3879  24836   .1562     1.2643 

        4  -10617.70  49582.00    0.7859   2949  24791   .1190     0.9629 

        5  -13282.40  48994.00    0.7289   2054  24497   .0838     0.6787 

        6  -10823.40  45982.00    0.7646   1504  22991   .0654     0.5295 

        7  -10031.00  38414.00    0.7389    933  19207   .0486     0.3932 

        8   -9408.70  28812.00    0.6734    520  14406   .0361     0.2922 

        9   -8910.30  19882.00    0.5518    247   9941   .0248     0.2011 

       10   -5007.70  12824.00    0.6095    141   6412   .0220     0.1780 

       11   -3255.70   7928.00    0.5893     66   3964   .0166     0.1348 

       12   -1516.10   2930.00    0.4826     21   1465   .0143     0.1160 

       13      96.90    338.00    1.2867      5    169   .0296     0.2395 

       14      -8.00      8.00    0.0000      0      4   .0000     0.0000 

       15       0.00      0.00    0.0000      0      0   .0000     0.0000 

       16       0.00      0.00    0.0000      0      0   .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       0.00      0.00    0.0000      0      0   .0000     0.0000 

 

 

My calendar year 2006 database broken out by JPRMLProb numeric value is shown below. Like the tote probability chart, this next chart should give you a clear sense of the JPRMLProb algorithm’s accuracy. The probability range for each row is indicated by the values in the min and max columns. The actual win percentage along with number of wins, plays, and flat bet win roi is also shown. For example, the top row shows data for horses where the algorithm said the probability of winning the race was between 0 and 5 percent. The win rate achieved for the 55,967 horses in that row was actually 3.46 percent.

 

     By: JPRMLProb

 

    >=Min       <Max      Gain       Bet       Roi   Wins  Plays     Pct   Impact

  -999.00       0.05 -38810.50 111934.00    0.6533   1936  55967   .0346   0.2800

     0.05       0.10 -22963.10  92706.00    0.7523   3497  46353   .0754   0.6107

     0.10       0.15 -14039.60  66112.00    0.7876   4064  33056   .1229   0.9952

     0.15       0.20 -11443.20  63086.00    0.8186   5347  31543   .1695   1.3722

     0.20       0.25  -2134.50  21634.00    0.9013   2437  10817   .2253   1.8237

     0.25       0.30  -2509.90  20606.00    0.8782   2693  10303   .2614   2.1158

     0.30       0.35  -1659.80  16286.00    0.8981   2525   8143   .3101   2.5100

     0.35       0.40   -737.10   7398.00    0.9004   1413   3699   .3820   3.0921

     0.40       0.45   -396.50   3308.00    0.8801    687   1654   .4154   3.3622

     0.45       0.50    -58.70   1272.00    0.9539    310    636   .4874   3.9455

     0.50       0.55    -39.50    284.00    0.8609     72    142   .5070   4.1043

     0.55       0.60    -10.20     54.00    0.8111     14     27   .5185   4.1972

     0.60       0.65      1.30      4.00    1.3250      2      2  1.0000   8.0947

     0.65       0.70      0.00      0.00    0.0000      0      0   .0000   0.0000

     0.70       0.75      0.00      0.00    0.0000      0      0   .0000   0.0000

     0.75       0.80      0.00      0.00    0.0000      0      0   .0000   0.0000

     0.80       0.85      0.00      0.00    0.0000      0      0   .0000   0.0000

     0.85       0.90      0.00      0.00    0.0000      0      0   .0000   0.0000

     0.90  999999.00      0.00      0.00    0.0000      0      0   .0000   0.0000

 

 

Bris Prime Power vs. QRating

 

Bris Prime Power

Much has been written about the Bris Prime Power Rating. It is a pretty good rating. For years it has been viewed as a sort of benchmark among software generated power ratings. In fact very few software generated comprehensive power ratings have been able to duplicate its win rate and roi. My calendar year 2006 database shows the following results when broken out by Prime Power rank:

 

     By: Prime Power Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -6749.10  50328.00    0.8659   7912  25164   .3144     2.5451 

        2   -7116.40  50750.00    0.8598   5096  25375   .2008     1.6256 

        3   -9084.80  50850.00    0.8213   3685  25425   .1449     1.1732 

        4  -11296.60  51400.00    0.7802   2750  25700   .1070     0.8662 

        5  -10256.50  50330.00    0.7962   2122  25165   .0843     0.6826 

        6  -12501.50  46618.00    0.7318   1424  23309   .0611     0.4945 

        7  -12865.70  38040.00    0.6618    912  19020   .0479     0.3881 

        8  -10811.80  27792.00    0.6110    517  13896   .0372     0.3012 

        9   -7103.70  18462.00    0.6152    306   9231   .0331     0.2683 

       10   -3618.30  11330.00    0.6806    167   5665   .0295     0.2386 

       11   -2095.50   5438.00    0.6147     74   2719   .0272     0.2203 

       12    -988.30   2542.00    0.6112     22   1271   .0173     0.1401 

       13    -269.90    556.00    0.5146      7    278   .0252     0.2038 

       14     -31.20    236.00    0.8678      3    118   .0254     0.2058 

       15      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       16      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       17      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       18      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       19      -4.00      4.00    0.0000      0      2   .0000     0.0000 

 

 

 

QRating

JRating and JPR aren’t the only comprehensive power ratings found in JCapper. JCapper2007 also has the QRating. Historically, each of the top four ranked QRating horses has outperformed the top four ranked Prime Power horses in both win percent and flat bet win roi. My calendar year 2006 database broken out by QRating rank looks like this:

 

     By: QRating Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -4323.50  49656.00    0.9129   7916  24828   .3188     2.5808 

        2   -6343.40  49660.00    0.8723   5150  24830   .2074     1.6789 

        3   -8006.70  49650.00    0.8387   3680  24825   .1482     1.1999 

        4   -9516.00  49604.00    0.8082   2789  24802   .1125     0.9102 

        5  -11757.50  48990.00    0.7600   2044  24495   .0834     0.6755 

        6  -12223.80  45972.00    0.7341   1421  22986   .0618     0.5004 

        7  -13316.40  38406.00    0.6533    890  19203   .0463     0.3752 

        8  -10418.90  28766.00    0.6378    546  14383   .0380     0.3073 

        9   -7921.50  19832.00    0.6006    300   9916   .0303     0.2449 

       10   -5791.70  12812.00    0.5479    153   6406   .0239     0.1933 

       11   -2906.60   6720.00    0.5675     69   3360   .0205     0.1662 

       12   -1752.50   3646.00    0.5193     34   1823   .0187     0.1510 

       13    -276.80    648.00    0.5728      4    324   .0123     0.0999 

       14    -234.00    310.00    0.2452      1    155   .0065     0.0522 

       15      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       16      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       17      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       18      -2.00      2.00    0.0000      0      1   .0000     0.0000 

       19      -4.00      4.00    0.0000      0      2   .0000     0.0000 

 

 

 

 

JPR (JCapper Power Rating) and Win Rate

JPR is a comprehensive power rating. Historically, the top ranked JPR horse wins less often than the top ranked Prime Power horse. But this is by design. When I created JPR I made a trade off. I traded win percent for roi. My calendar year 2006 database broken out by JPR rank looks like this:

 

     By: JPR Rank

 

     Rank       Gain       Bet       Roi   Wins  Plays     Pct    Impact

        1   -4201.90  49656.00    0.9154   7401  24828   .2981     2.4129 

        2   -7015.80  49658.00    0.8587   4982  24829   .2007     1.6242 

        3   -8186.00  49658.00    0.8352   3676  24829   .1481     1.1984 

        4   -9533.10  49604.00    0.8078   2781  24802   .1121     0.9076 

        5  -12419.20  48986.00    0.7465   2150  24493   .0878     0.7105 

        6  -13247.30  45970.00    0.7118   1569  22985   .0683     0.5526 

        7  -12498.10  38410.00    0.6746   1078  19205   .0561     0.4544 

        8  -11488.90  28770.00    0.6007    641  14385   .0446     0.3607 

        9   -6677.70  19828.00    0.6632    372   9914   .0375     0.3037 

       10   -4872.70  12810.00    0.6196    212   6405   .0331     0.2679 

       11   -2885.10   6718.00    0.5705     86   3359   .0256     0.2072