Toteboard Predictions

[Ray2000]

It's amazing how a collection of gamblers, robots, horsemen and bowling party goers accurately predicts the chance of winning for each horse in a race. Below is a graph showing the actual percent winners (in 20 odds ranges) vs the crowd's prediction. There are 10,000 starters in each odds segment, 17% takeout.

The bettors slightly under bet the heavy favorites (overlay) and way over bet the extreme longshots.
But they're right on with the 10 percent 'chance to win' starters.

[Capper Al]

Would you explain this a little more?

Thanks

[Ray2000]

I looked at Harness race final odds for 2014-2015 and grouped the starters into 20 odds ranges.
The midpoint odds of each sector was used (along with an estimated 17% take out) to show what the 'crowd' believes the chance of winning is for this odds group.

% chance = (1-take)/(Odds + 1)

I then compared this to how many horses in that odds group actually won their race. The crowd is most accurate in the 5/1 - 7/1 range.

I also looked at what kind of return would you get if you bet every horse in that group.



Odds Ranges Horses Winners Win % ToteB $1 ROI %Return
>=0.05 <1 10690 5995 56% 52% $9,361 88%
>=1 <1.6 10995 4277 39% 37% $9,612 87%
>=1.6 <2.15 10731 3229 30% 29% $9,156 85%
>=2.15 <2.8 10921 2729 25% 24% $9,384 86%
>=2.8 <3.5 10834 2145 20% 20% $8,771 81%
>=3.5 <4.3 10918 1842 17% 17% $8,897 81%
>=4.3 <5.2 10938 1619 15% 15% $9,212 84%
>=5.2 <6.2 10225 1218 12% 12% $8,077 79%
>=6.2 <7.5 11481 1226 11% 11% $9,514 83%
>=7.5 <8.9 10518 893 8% 9% $8,139 77%
>=8.9 <10.7 11273 818 7% 8% $8,743 78%
>=10.7 <12.8 10811 621 6% 7% $7,847 73%
>=12.8 <15.4 10913 539 5% 6% $8,062 74%
>=15.4 <18.6 10648 416 4% 5% $7,403 70%
>=18.6 <22.9 10912 347 3% 4% $7,457 68%
>=22.9 <28.5 11022 268 2% 3% $7,048 64%
>=28.5 <36.1 10888 208 2% 3% $6,805 62%
>=36.1 <46.9 10868 163 1% 2% $6,844 63%
>=46.9 <65.2 10855 94 1% 1% $5,250 48%
>=65.2 10881 40 0% 1% $3,573 33%


Hope this helps

[Capper Al]

Thanks. That clears it up.