Bandini would have gotten a slightly better figure if they had both been at equal weights, but High Fly spotted Bandini 4 pounds.

The other question is more complicated, and interesting, and gives a good insight into how we make figures. First of all, I don't look at the Thoro-Patterns when making figures for a race, but I went ahead and averaged the percentages of the horses going INTO the race (different than the ones that appear on the sheets now that they ran what they ran). In the "new top" category (meaning, more than one point better), it came up just under 15%, and just under 30% for the "pairs" (meaning within a point either way). It would certainly be within the normal range if anywhere from zero to three horses ran new tops, but one out of the nine happens to be the closest you can get to 15%, and 4 pairing tops (44%) is certainly close to the norm.

But that's not how we do it. Leaving aside the surrounding races and any other info, what we are doing is looking at various scenarios, to see which fit the best. Try any other way to do it, and you'll see what happens-- the only other one which makes any sense at all is to take off two points, which pairs High Fly and Park Avenue Ball, but if you do that you will have 5 of the 9 horses (55%) running new tops. Statistics aside, logic tells you that it is far more likely that a group of horses (as opposed to ONE horse) would run almost exactly back to their previously established maximum level, than choose the exact same day to improve on that level by almost exactly the same amount.

Interesting question. I guess it's fair to say that we think in those percentage terms in an intuitive way, but it would be wrong to apply them rigidly to small population groups. Using it as a check on LARGER groups is another story... hmmm.

Attached find the sheets (and Thoro-Patterns) going into the race. To see how they ran in the race, see the first post in this string.

http://www.thorograph.com/hold/foy-pre2k5.pdf




TGJB