What better to get stuck into than a fair few non-winning selections? As a preface I'll take the opportunity to stress that the ultimate concern in my thinking is finding the likeliest winner of a race. In that respect backing three, four, five and more horses in any one race makes no appeal - that in itself is very easy to do. Trading holds no place whatsoever in analysis either. The other point of note is the general ethos behind winner finding: how accurate is the analysis trying to be, how predictive and how intent on reading the race correctly or more seeking a pay-out / ROI.
The strongest conclusion (as I've already thought about this a little) is likely to be that the process is sound (this will require mentioning successful bets from BST from last season but then this blog is merely an extension of that anyway) but has structural deficiencies arising from a) the timing of race analysis due to work commitments; b) variance; and c) a lack of focus on ROI. The latter one is quite surprising but it seems to be the case that initial financial return of any kind has been overlaid by backing a singular strand of analysis. All that should become clearer as I explain my losing bets posted on here; anyone reading where I went wrong AND how I was unable to steer myself onto the correct path may derive some benefit for their own approach.
Paddy Power Gold Cup
Perhaps one the trickiest things I find is weighing up the absolute quality of a race. On this occasion it wasn't a problem. It was a shocking renewal on desperate ground. I had little interest in the race having initially identified Notus De La Tour (more on him later) and with activity elsewhere focus on the race was below the required standard (#error). Because of the poor nature of the race in absolute terms (I did not rate Walkon highly at all before the off) it seemed clear that if Grand Crus reproduced his Feltham win he was by far the likeliest winner. A breathing problem (at least) ensured he was pulled up. The main #error was approaching the race from a very early point of view. Last season and indeed whenever possible decisions were left as late as possible. A combination of work commitments and poor overall focus led me to tackle the race early. This caused two problems: firstly, having decided on the shape of the race before confirmed runners and ground were known, I was blinded to the late swell of activity and betting moves for the winner Al Ferof. Secondly this meant I was unable to give Al Ferof due consideration and instead relied on an overall impression of him from last season. Whilst this is still probably accurate (I think) it meant he was not assessed in the context of this race and its super-weak field. This led to an improbable u-turn on the fav - whom I had raised considerable doubts about on these pages - due to the initial weighting of the race as poor and there being almost no alternatives (this in itself is often not the case of course). Whilst I didn't back Grand Crus for the race outright (he went into some small multi's instead) the lack of focus and desire/need to play the race early led to a poor conclusion.
Hennessy Gold Cup
The above was compounded by a classical timing + analytical error in this race. In essence, this race boils down to Graded class ability v handicap (progression) ability. With a deluge of rain mid-week and the favourite Bob's Worth perhaps unlikely to run in such conditions (or if running on a bog) I again played the race ante-post and again it was unsuccessful. With deep ground the call was made to side with handicappers who had shown fair form and who might operate best over the trip. In terms of actual finishing positions the selections were good: taking away the top three who were all Graded class they finished in second and fourth (fifth & seventh). Whether the ground was in fact ever a concern for the analysis (class outs) is a kind of mute point as the ground dried significantly and the three Graded class horses ran through the deep end of the race to the line. With day of race conditions the decision not to side with fair handicappers would have been much easier but even here we have a key #error: "In essence, this race boils down to Graded class ability v handicap (progression) ability." so the correct play would have been one of each, a handicapper and one of the classy crop. Given I have continuing doubts as to the residual class and resolution of First Lieutenant then the winner could have been included in a strategy designed on ensuring an ROI. Strange as it sounds, I was happy to double-barrel the handicappers rather than take a more balanced approach which has to be noted for future runnings where the final decision is deemed marginal (which it was here during the downpours). This is also linked to the Likeliest Winner approach as covering all race scenarios leads to unwarranted numbers of bets on several horses in one race. The handicap picks were good; the approach to the race balance was not, ultimately leading to a poor conclusion.
December Gold Cup
Much less self-criticism required here as this was more than anything simply variance given my overall approach. A strategy I am very happy with is to rarely ever back 4yo's in big(ish) handicaps and to never back horses aged 12+. It simply rules out a lot of horses that return a miniscule win ratio. Of course there is a contra point that surely it is best to assess races on their individual merits and work from there and that is viable but I had already had a reversal in this race with a 5yo (as mentioned, Robinson Collonges) and despite Unioniste's "tougher" profile he was still a baby in my eyes! That overall explains why I said he "should be a million". Of course were it solely that it would be a slightly enigmatic approach but there was of course a very interesting candidate on paper that reminded me a little of Great Endeavour and it was a shame that Notus De La Tour developed a physical problem or just wasn't right on the day. Walkon has now finished second in both big intermediate handicaps. Analytical flaw? Quite the opposite and I am very happy to always field against this type of animal, for all that he has actually run quite well! He just hasn't looked like winning either race and for all the media coverage and shortness of price his profile is just neither progressive nor does he have performances of residual class in the bank following his time on the sidelines when injured. There is a good post on BST (not by me obviously) that shows the potential benefit for those who took Unioniste on merit rather than age. They'll win bad races like that now and then but little else outside of their own age group at that level.
The Betfair Chase was a quietly compelling race that was fine analytically with the winner pin-pointed. The real benefit was, potentially, doubling up the winner with Long Run in the King George. This is part of an overall strategy of pot control this season - to not get caught up in the here and now of any one race but also to maximise correct thinking across a number of races for a bigger return later down the line (should such longer term thinking bear fruit). This example gave 14/1 about Long Run on Boxing Day. The downside is not maximising the on-the-day win but in terms of the races highlighted here we can see that that's not a problem! Had they won though the potential for greater returns would kick in which is something I'm happy to pursue this season for a number of strategic reasons.
The Becher Chase saw another selected runner pulled up (this time with an abject display) but as a returning hero this is again more variance than anything else, particularly as the winner was a 14yo! West End Rocker remains well handicapped in part because he is just so impossible to be consistent with. Well, he may have the odd mental issue too. As an overall strategy he goes hand-in-hand with other returning heroes like Merigo and Always Waining and as such they will not always repeat win and this was a case in point. In theory the process was solid because had he performed then based on his demolition of Niche Market last season he would have accounted for a field led by the admirable but now retired (!) Hello Bud.
The fact that the four handicaps failed to produce a winner, whereas the Grade 1 contest proved accurate has personal ramifications. Handicaps are often best approached as late as possible when all information is known except in those circumstances where a horse has been tracked specifically with a certain race or conditions in mind. Handicaps in particular often take on different complexions as stable information is slowly revealed and certain runners come and go, with ground conditions assured (many times they are but this season has provided a good example of rapid change from stated/suspected descriptions).
In contrast the abilities of graded horses can be more accurately defined from some way out: intentions are often clearer, preparations much more significant and traceable. In the Betfair for example, from a long way out it was made clear that Long Run would have a gentle introduction and that Silviniaco Conti was race-fit and firing on all cylinders for the race making assessment easy.
Focus, patience and tactical betting all require further work with a keener eye kept on ROI policies, strange as that may sound: without an A-Game mentality analyses can wend a winding path leading to the kind of errors identified above.
No comments:
Post a Comment