Quote:
Originally Posted by ARAZI91
Hi Steve
Yes the times are official taken from the PF (Racetech is the operator) and already converted by the lengths per second scale as stated used by the BHA based on the official going (usually done by readings from the Turftrax Going Stick) - https://www.turftrax.co.uk/goingstick.html
I have attached the Lengths Per Second scale used here , although as i said previously through my own studies i have noticed inconsistencies from what they have stated they "should" be using to what they have actually used.
On Flat Turf the scale should be 6 lengths per second (0.166) on Firm ground to Good ground , 5.5 lengths per second (0.181) on Good To Soft and 5 lengths per second (0.20) on Soft / Heavy. They have separate scales for All Weather Racing and National Hunt Racing.
I basically replicate these conversions so that my figures are in line with the BHA's Official Handicap ratings , as these are a big part of my method of rating. |
if you have the times then you need not worry about margins or seconds per length or anything.
if you have times for all runners then what more could one ask?
the times will mean you don't need to worry about anything those official guys say on the subject, you can ignore them and be better off for doing it.
forgetting weight(mass & hcp) and pace and all the extraneous stuff for now.....
time is just about percentages/ratios, nothing else.
if you win a 1000m race in 58 and a beaten runner does the trip in 59 then assuming the race speed was 100 then it's just 58/59 = 0.983050847
if the race was 2000m(and again assuming 100 speed for race) and they run 124 and a beaten one runs 125
then it's 124/125=.992
now i am being a bit precious here because one of them is wrong, and it's the way of beyer(even if he does not realise it himself)
the above are just beyers but expressed as simple percentages.
to understand them better then......
0.983050847 needs to be converted to .....of a 1000
(0.983050847-.9)*1000 = 83.05
so if the winner was 100 then the beaten one is 83.05
for the 200m race....
(.992-.9)*1000= 92
now lets just assume there was another race of 1000m and 2000m on the card and they were for jeju ponies!the 1000m was won 64s and the 2000 in 130 and each race had a runner beaten that same 1 second.
64/65= 0.984615385 = (0.984615385-.9) * 1000 = 84.6
130/131= 0.992366412 = (0.992366412) * 1000 = 92.37
thus beaten a second in 58 second race is not the same as being beaten a second in a 64 second race.
ditto for the 2000(which is still wrong in both cases, as would be beyer)
now let's compare the speeds of the four races....assuming the first two were run equivalent to the standard which would give them a speed of 100
if 58 equals 100 then 64 equals.......58/64=.9375 = (0.90625 -.9) *1000 = a race speed of 6.25....i said they were jeju's!
if 124 equals 100 then 130 equals 124/130=0.953846154 = 53.85
in this case the race speeds would be correct even though using the same way for beaten runners is incorrect.
it's just a long winded way of saying raw times are about ratios and not much else.
it also means that you don't need any constants.
everything is proportionate.
no assumptions and in my view making asumptions when none needed is bad.
how the beaten runners should be handled is only slightly different, and you need to account for the race distance in your calculation but is basically the same....any takers on how?
EDIT: just needed to fix some errors.....too early for wine stephen!
hopefully there is no more mistakes.
