August 12, 2006 at 12:19 #2864
<br>Don’t put much value on race trends myself, but it’s interesting stuff your posting.
Doesn’t Andrew Mount do a similar type of analysis with his Trend Horses?
(Edited by empty wallet at 1:19 pm on Aug. 12, 2006)August 12, 2006 at 12:28 #75229
Trends are the most under-rated aspect of race evaluation IMO.
I agree that trends can be meaningless when used in the crude fashion they often are but, used intelligently and creatively, trend analysis is a very powerful tool.
I like the approach EC has outlined above. To have an edge you need to think differently. For example, how many people involved in establishing market prices on betfair for this race have considered the foaling date? Very, very few I’d imagine and it may therefore give EC (and the rest of us now!) the little edge that is needed.August 12, 2006 at 12:40 #75230
Quote: from EC on 1:22 pm on Aug. 12, 2006[br]form breaking is a little different i reckon…more realted to the horse of course..instead of the race<br>
Yes, apologies EC, Trend Horses is horse analysis and not a race trend analysis
(Edited by empty wallet at 1:41 pm on Aug. 12, 2006)August 12, 2006 at 12:40 #75231apracingParticipant
- Total Posts 3105
<br>I’m certainly a believer in the value of trends, but would just offer a word of warning about using draw stats on the July course.
Given there are actually two parallel 8F courses that behave differently in terms of any draw bias, you need to know which track was being used for past runnings in order to be confident about the trends.
Like EC, I look at foaling dates for 2-y-old races and I also consider the stamina index of the sire in races at 7F and upwards.
Having said that, rain on watered ground is putting me off any involvement at Nmkt today anyway – who knows what the jocks will decide to do?
APAugust 12, 2006 at 13:56 #75232
If a trend or bias has an effect then it will have an effect on beaten horses as well as winning ones. Using just winners is inadequate.
"Filtering" (e.g. "rule out 4yos", "rule out horses beginning with the letter S", "rule out horses who have not won in the past year", that leaves only two for the shortlist) treats all criteria as if they are of equal value.
A horse that has many trends apparently strongly in its favour should not be ruled out on account of failing on one count of questionable significance. Trends should not be treated as if they are a simple binomial.
More than anything, however, most trends deal in generalisations, but it is specific horses who win (and nearly win, and lose) races.August 12, 2006 at 14:09 #75233
Goinng against yer trends EC
Precocious Star is the value bet iimoAugust 12, 2006 at 14:13 #75234
Nice win for the trends! Nice work EC.August 12, 2006 at 14:24 #75235
Pru – You are right in asserting that it is specific horses who win (and lose) races and that trends deal in generalisations. However, it is the case that certain races (not all races) do throw up results which do follow ‘general’ trends.<br>It is the long term results which are important and, while often a ‘specific horse’ will buck the trends, following strong generalised trends can and does pay off if used carefully.<br>When examining trends I never look only at winners. It is imperative that the credentials of losing horses are examined also.
It is no good knowing, for example, that 80% of the winners of a particular race came into that race having won last time if, when you look at the losing horses in that race, you find that they shared that profile. You are, effectively, aiming to establish characteristics in the winners which separate them from the losers in the history of a race.August 12, 2006 at 14:27 #75236noreMember
- Total Posts 151
nice work, ecAugust 12, 2006 at 14:29 #75237
<br>Can yer put it use it on next race at Ascot and tell me if drop back to 12f will suit Mikao or not?
I’ll go skint following him btw :biggrin:August 12, 2006 at 15:31 #75238
Another thing about trends is that you need to be careful about double counting. You may discover that 3yos have had a good record in certain all-aged handicaps at this time of year (and I’d say that that could well be a significant, though not over-riding, discovery) and that lowly weighted horses have the same.
As 3yos are more lowly weighted in general due to their immaturity, the two are in fact related.
I use average cumulative lengths beaten, with a stop-loss to prevent tailed-off horses from skewing the data.
I look at "trends" more now than I used to (not entirely through choice), but a trend would never put me off backing a horse that goes against it if I thought its specific claims were strong enough.August 12, 2006 at 15:41 #75239
"I look at "trends" more now than I used to (not entirely through choice), but a trend would never put me off backing a horse that goes against it if I thought its specific claims were strong enough."
Agreed.August 12, 2006 at 15:52 #75240
For once I am at a loss to know what to say: I am only used to dealing with hostility and indifference!August 12, 2006 at 15:55 #75241
Ok, here goes.
Foaling dates – referred to earlier on this thread – make only a small difference in general. If it appears otherwise in specific instances it could well be just a statistical blip.
(Edited by Prufrock at 4:55 pm on Aug. 12, 2006)August 12, 2006 at 16:45 #75242
Well, just to introduce a little hostility back into it, I only highlighted the foaling date issue as an illustration of the type of area which could potentially yield an edge.
Despite my apparent championing of stats and trends they can quickly result in one becoming bogged down and, eventually and inevitably, consumed in a quicksand of irrelevances and meaningless numbers (‘blips’ being only one of many types of quicksand on this most treacherous of beaches) unless continual efforts are made to ensure that you are focusing on relevant material.
There is one critical test that will starkly illustrate that.
The profit/loss figure!
- You must be logged in to reply to this topic.