Those two words in combination are almost taboo these days. The mere mention of Derby stats tends to evoke a scoff, a raised eyebrow and a retort about how Barbaro and Funny Cide abolished all those useless stats. The fact that stats may weigh into your Derby handicapping makes some people believe that you're an out of date dinosaur still holding to old methods that have stopped working. I would submit that those people both do not fully understand the difference between a good statistic and a bad statistic and that they do not realize the power of good statistics.
In my field, which is trading securities, the ability to compile statistics is pretty common. The ability to determine which ones are viable is not such a common skill. Just because it seems to work doesn't mean its a viable statistic. What makes a stat viable and useful is a sufficient sample size and a quantifiable reason why the statistic should work. In other words no jinxes or myths. Had Barbaro's mother been flown to Montana to foal him he still would have been Barbaro and he still would of been capable of winning the Derby. The existence of so many bad statistics is the real reason why so many people are skeptical, and well they should be. But it also creates an opportunity for those who can identify the god and weed out the bad. It is my belief that every single factor must be aimed at identifying six key ingredients to winning: Speed, Fitness, Current Form, Class, Experience and Suitability to the conditions.
I have been compiling Derby statistics for many years now and I have been working at building a profile based on the most relevant of statistics. I call the project the "Triple Crown 20-20" partly because 20-20 represents clear vision and partly because 20 is the perfect profile score than can be achieved. The basic premise of the project is that there are 20 statistical factors that I deem to be important for Derby success. I run every contender through that grid and they are given a score based on how well they fit the profile. I haven't made my 20 factors public yet but to see the profile scores achieved over the last 11 runnings you can visit the link below.
Triple Crown 20-20
As you can see the profile has worked quite well in identifying winners. It has also not done too badly identifying the horses who fill the minor placings. A simple $2 win bet on each horse with a perfect score over the years would have netted you $179.80, a 428.10% ROI, and that is even without Giacomo. In fact if you simply bet every Derby horse over the last 11 runnings based on their profile score (ie 20pts = $20, 14pts =$14, 6pts =$6) you would have made about 25%.
Every Wednesday from now until the Derby I will seek to share another piece to the Derby puzzle from a statistical standpoint. Either sharing a relevant statistic or debunking the myths about bad statistics.
Subscribe to:
Post Comments (Atom)
3 comments:
I'm guessing that running in large fields isn't one of your factors since Bro Derek got a 20, but I think now more than ever, having experience running in large fields is vital. If a horse hasn't been bumbed and steadied in a prep, then they aren't prepared for the 20 horse pile-up that is the Kentucky Derby.
i read with great interest KC's, 'derby statistics' post//while i am not good with statistics, i do think good statistics provide consistency and success in handicapping/ being a weekend warrior and a punter so to speak, i actually want the reason i select a runner to be a sound one because generally, it leads to other successes // i had last year's derby trifecta for the first time in many years / i don't mine data- bases and statistics like the professionals use but, try to use past situations where something is likely to happen// much like one of the approaches used by andy beyers// hopefully, the tremendous payoff i received was based on sound reasoning/// i remember reading, apparently a true story back in the days of pennsylvania phil, about a better who bet and won on a 50-1 shot because the rider loaded in the gate with both feet out of the stirrups/ he thought this was some kind a signal to someone that the horse was going to win//as it turned out he was able to ask the jockey and trainer about it later and was told the legs out of the stirrups had nothing to do with anything//and so it goes///i am interested i KC's definition of 'fitness' and 'suitability to the conditions'//also, i was wondering how thunder gulch and sea hero might have faired in KC's triple crown 20-20/// chicago (go bears) gerry
"i was wondering how thunder gulch and sea hero might have faired in KC's triple crown 20-20"
I cannot say for sure because I lack complete data for the entire fields. Some factors depend on who the horse is facing. That is why the profiles do not go back past 1996 but my estimates are that Thunder Gulch got a 16, Go For Gin got a 20 and Sea Hero got a 6.
Sea Hero is essentially the worst Derby winner of modern times.
Post a Comment