[Ted Craven]

The sharp move you cite and your race graphs reminds me of the old 'Pace Vector Graph' which plotted horses' velocity gains against the pace of race, though that graph showed lines selected from horses' various past races - and your graphs are much easier to see the moves within a single race.

With respect to only the 3 examples you posted here, how about this approach:

1. for a race today, scan horses' PPs for a position gain of a specified magnitude in their last (or 2nd back?) race - e.g. 5 positions from 4f->Str (Southern Arch), 4 positions in 1st Fr (Real Tipsy), 3 positions from 4f->Str (Skewed) - and perhaps where the gain occurred is important

2. when found, check the result chart for that prior race and calculate a 'vector graph' of gains against the POR (measured by velocity or time), looking to see if position gain also translates to 'standout' vector gain (by some definition), and measuring how forwardly placed the gain was (e.g. Southern Arch's was more significant than Miss Smoothness who was not in the hunt), and also perhaps under how much 'pressure' the gain occurred, e.g. by measuring how tightly clustered the beaten lengths were at the ending call point

3. if the horse in question then faded (due to the exertion of the gain, or the pressure during the gain) or did not fade (Skewed) and today is shorter

4. then you have a markup to identify such horse in its race today and observe more closely versus the rest.

To clarify, I would define a computed 'vector gain' as the positive difference between the velocity of a horse during a split and the velocity of the Pace of Race during the split (e.g. F2_POH - F2_POR) - then compared to the other horses vectors.


Can you tell me the next race, for each of Real Tipsy and Skewed?

Thanks,

Ted