The 'Second String' Effect
|
|
The 'Second String' Effect |
|
|
Article 12_002_1
10th January, 2012 |
Whenever possible the articles in 2012 will try to link back to the 'Pointless Punditry' piece of 2011. On mature reflection that discussion article raises some very important issues and there are no signs that the problems will fix themselves unaided. The aim of this article is to consider how much 'The Market' really knows when it comes to 2yo racing. Specifically, for consideration here, how tuned in it is when trainers and owners have multiple runners in a race?
The sections below take examples from the 2010 and 2011 seasons for major connections. The purpose is to see whether 'The Market', and by extension the professional pundits, has a good handle on which of the multiple runners are likely to perform best. 'Professional' meaning those pundits employed by others and not those betting for themselves. The Hypothesis to be considered is this :- The Market formers get idle and sloppy when faced with lightly raced and unraced 2yos. Lacking even the limited crutches they normally rely on they overvalue indicators from other sources and allow others to think for them.
In a 2yo race where a major trainer has multiple runners there will normally be a known heirarchy of jockeys that the stable uses. In the cases where owners have multiple runners the jockey angle will be added to by the owners having to nominate which horse will wear the 'First Colours' and which the second. The Market formers will take these indicators as unequivocal confirmation of the quality of the runners. It will assumed that the trainers and owners know perfectly how well their representatives will perform and that the second strings, or third and so on, can be discounted as winning prospects.
The B2yoR view going in was that the way the Market reacts to multiple runners can be characterised as a 'Second String' effect. They can end up at prices which are too long and do not reflect properly the real chances of success of these horses. Over a longer period the Second Strings will prove to be good value and profitable to follow in the right circumstances. The following sections consider a number of example to see whether there is any worthwhile substance to support that view.
[April 9th Update - Note that this article was written before the changes of jockey which occurred in March 2012. Richard Hills retired as Hamdan Al Maktoum's retained first jockey before the 2012 season but remains closely involved and intends to still ride work. Paul Hanagan took over as Hamdan's principal jockey but will try to ride for Richard Fahey whenever possible although Tony Hamilton seems to be the first pick for the Fahey stable when Hanagan in elsewhere. But, these are probably details which need to be assessed in the context of the message of following sections rather than changing the underlying effect.]
To start the overview the Richard Fahey stable is a good example with a large numer of 2yo runners each year and an above average record overall. Fahey appears to prepare his juveniles quite fully for debut and has produced a well above average first time out (FTO) strike rate in each of the last six year through to the end of 2011. You would have made a profit backing all of his 2yo debuts in four of those seasons. The stable has a range of jockeys attached to it and a cast of regular work riders and many of both groups have been with the stable over many years. Champion jockey Paul Hanagan is clearly the 'Top Man' in the set-up and you would expect him to have the first pick in any race where Fahey has more than one representative.
The survey will limit itself to Maiden and Auction races where the most lightly raced horses should be found. The table below summarises the results from those types of races in 2010 and 2011 where Fahey had multiple runners.
| Season | Number of Multiple Races |
Num 1st String did Best |
1st String Wins, SR & P&L |
Num 2nd+ String did Best |
2nd String Wins, SR & P&L |
| 2010 | 18 | 9 | 3, 16.7%, -10.18 | 9 | 1, 5.9%, -1.00 |
| 2011 | 23 | 16 | 5, 21.7%, -1.5 | 7 | 3, 12.0%, +35.5 |
An interesting story emerges from the table and not one B2yoR was expecting, unlike some of the later examples. At the top level the table shows a healthy number of instances where the longer Starting Price runners (usually not Hanagan ridden) has done best. Although only producing 4 wins the 'Second String' groups SPs have been at relatively larger prices. In 2010 the single winner was at 16/1 and produced a small loss with 18 'Second Strings' that year with two races with three Fahey juveniles in.
2010 also saw Second Strings place at 20/1 and 14/1 doing best of the stable representatives so not that far away from a solid win profit. Looking into the backgrounds of those two places is instructive. Both finished runner-up in essentially empty Northern races behind a solid winner from another major stable (Smart & Johnston). If the solid quality winners had not been in the race or had underperformed then the rest of the fields were very poor quality. In those set-ups no usable horse from the Fahey stable should have been at those long SPs. However, both of those placers were still maidens at the end of their 3yo careers and so were the Fahey runners they were Second String to.
The elements to pull out of that include two possible important points. Firstly never underestimate how weak some Northern races are. Secondly, even if you believe a trainer is competent in assessing the ability of his own horses remember that they, and hence the Market, do not fully integrate that knowledge with the ability of the rest of the race field from other stables. You can imagine Fahey saying that both those second string placers were limited types and he was not expecting much from them and they start at long SPs. He would be as surprised by the places as anyone but with 18 months of time passing and them both still maidens it is clear he was right in his assessment. Where he was 'wrong' was that he thought a solid quality (say OR65+) effort would be needed to place.
Move on to 2011 and Fahey had three Second String winners at 9/2, 20/1 and 33/1 with the last of those as Third String behind horses at 6/1 (Hanagan ridden) and 12/1. In this example the 33/1 seemed to be the best of a moderate trio he ran in one race who were all on debut. The other pair have not managed wins as 2yos although the Hanagan ridden Colbyor managed places in moderate maidens. You presume another weak Northern race and the 33/1 a real 'Second String' effect and not properly reflecting the relative abilities of the three and therefore a value price.
The other two winners were OR75+ fillies when Fahey had the short priced favourite beaten, ridden by Hanagan. In the case of the 9/2 win Listed quality Lily's Angel was on debut in early season while the odds-on favourite was One Kool Dude who had run placed on debut in Ireland for the stable. While Lily's Angel was going on to win at higher level One Kool Dude ran another 9 times without winning as a 2yo. After the race Fahey said something on the lines of "... I feel sorry for Paul in that we knew Lily's Angel was alright but we thought One Kool Dude would be readier for the day...". If you want to believe the Market, and those that inform it, know everything you can produce a list of areas from the quote to make that belief seem misguided, at best.
But, if you are looking for evidence that the Second String effect exists the good profit from backing all the Fahey non-First Strings in 2010-11 would be encouraging. Not regular winners but long SPs, at value level, when they do occur. With which to move on to a more difficult trainer to study in Richard Hannon. The sheer quantity of runners that Hannon has means a much higher level of multiple runners and sometimes with his preferred jockeys not available. Hannon does not really target FTO wins except with his best runners which often means mixed results when he has more than one debut runner in a single race. The table below summarises 2010 and 2011 for Hannon.
| Season | Number of Multiple Races |
Num 1st String did Best |
1st String Wins, SR & P&L |
Num 2nd+ String did Best |
2nd String Wins, SR & P&L |
| 2010 | 77 | 47 | 19, 24.7%, +22.08 | 30 | 6, 7.0%, -4.00 |
| 2011 | 94 | 64 | 20, 21.3%, -8.32 | 30 | 7, 6.3%, +39.00 |
Backing all of his Second Strings (and higher) produced a small loss in 2010 on a total of 86 bets and a good profit in 2011 on 112 bets. The profit year including a very surprising 66/1 win for a second string at Ascot. On one hand you might say that this was so unusual that it should be discounted for long-term thinking. The counter view would be that this horse starting at 66/1 is exactly the sort of result that a 'Second String' approach is waiting to fall into.
When Ortac Rock won at Ascot he was on debut and ridden by Adam Kirby on what might have been his first ride for the stable. Hannon stable attached jockey Dobbs was on Eurytheus who was on his second outing but at 28/1 after an anonymous debut. Although he started out at what must be Hannon's favourite track in Goodwood so likely to be a usable type and he proved a solid maiden winner next time out. Ortac Rock went on to win a nursery off OR82 second time and runner-up in a major nursery after that off OR86 when supported as 7/2 favourite. Since he was at least the OR80+ type that Hannon typically starts at Ascot why was he ever at 66/1? The Second String status and the jockey presumably had some effect. Although the SP returns for that Ascot race look odd in other ways. For example, the 66/1 runner-up was trained by John Gosden, won a maiden second time up before having a go at Group level. Again, that horse was the second string to the Gosden trained favourite, also on debut, owned by Khalid Abdulla.
Looking at another specific example of a second string winner in 2011 links to a couple of issues which may mean the Market will struggle to weigh up the chances that well. Hannon starts a lot of good types at Goodwood and has had 10 debut wins there in maidens in 2010-11. In early June 2011 he ran Coup De Ville and Chandlery in a 6f maiden at the course. Jockey Hughes rode Coup De Ville who vied for favouritism but ended at 4/1 in a field where runners from other trainers were talked up. Chandlery was ridden by Dobbs and at 11/1. Both horses started slowly but while Dobbs got Chandlery organised quickly and settled in second place. Hughes was in his 'we have to teach everything to settle..' mode at the time so sat further back. Which meant that the strong finish Coup De Ville produced could not get him past Chandlery. The first home went on to win at Group level and the second won four times up to Listed level.
After the race Richard Hannon said something on the the lines of '.. I feel a bit sorry for Hughesie because I told him they were both nice horses but I thought Coup De Ville would be sharper today. CDV is trained at our second yard (at Herridge) and Hughesie wont have ridden most of the horses there at a busy time of the season like this.'. The point to pick out of that is that even those most closely associated with the stable may know less about the horses than the Market likes to think. Also, that even Hannon's famous 'Visual Dial' eyes can get things wrong even after 40 years.
The second point to note is about jockey's styles. Go back to the 2000-2008 period and spotting which Hannon juvenile was the 'First String' used to be tricky. Hughes was retained by Khalid Abdulla so did not have first pick with Hannon regularly. Ryan Moore was there as the stable apprentice and then remained associated after losing his apprentice claim. He seemed to have at least as strong a call at points. Add in that Dane O'Neill and Dobbs were also around having been apprentices at the yard and things were difficult. You then had to add in how likely each jockey was to be able to deliver a horse on debut to compete to win. Hughes would probably rank 3rd or 4th on that list and Dobbs as an apprentice more likely to get a FTO runner home.
With Hughes losing his retainer with Abdulla, Moore becoming Stoute's first jockey, O'Neill moving on to Henry Candy's things have seemed to be clearer in recent times. Hughes was definitely the first choice now and Dobbs the second choice. But, as the Chandlery race showed that still does not mean he will choose the right horse and deliver it to beat the Second String every time. The Market may well have over-reacted to the Hughes 'choice' and the Second String a good horse at a value price. You could also consider that Hughes may be riding to develop a horse on debut and not go all out for the win as another factor. The final point to note would be that when Hannon uses another major jockey, usually Fallon or Fortune these days, in races with multiple runners they will often be on a solid maiden winner type or better. Again, they will be accorded 'Second String' status by the Market but may well be equally as competitive to win as the Hughes partnered runner.
As a final example in this overview section let us consider owner Hamdan Al Maktoum. He has horses with a variety of trainers spread across the country from John Dunlop on the south coast to Mark Johnston in North Yorkshire. He retains two jockeys with Richard Hills the long time Number One and Tadgh O'Shea now the second choice. Which makes the 'problem' for Richard Hughes of not being able to ride all the juveniles available to him in their work seem trivial. Hills will probably have never ridden the majority of the Hamdan juveniles in recent work before they make their debuts. He will be relying on the trainers' abilities to judge their horses and probably filtered through Hamdan's racing manager. In the spirit of the quotes above second jockey O'Shea said after riding a Hamdan older horse second string to victory in a major race ".. it is hard for Richard to always make the right choice, he will not have ridden many of the horses recently, if at all,..".
Another point to note in this set up is that the Second Colours do not always mean the less fancied horse. Angus Gold, Hamdan's racing manager, has said explicitly at times that the choice of colours for multiple runners to wear does not always reflect their expected performance. He says that other factors come into the choice. For example, an owner bred horse out of a favourite broodmare of Hamdan's may be preferred over one purchased at the sales regardless of the ability they may have appeared to show in work. Add to this that the trainers have varying abilities to judge their horses and things can get a little confusing. A trainer like John Dunlop seems to struggle to assess his own horses at home and there are several examples over the years of his own second string winning. Never mind comparing his horse to one from another stable. The classic example being the 40/1 win for the useful Bunood in 2005 ridden by second string jockey Supple while Hills was on the third home as 7/2 second favourite.
The following bullet points summarise the performance of the various 'strings' in 2011 :-
An overall point from that list is why have 33 horses produced only 5 winners during the turf season and the 'First Strings' no more convincing than the others? Only one win for a Second String but again the SP was long enough to ensure little harm and give hope that a larger sample might produce a profit. The performance of the Third Strings in later races is a little baffling. You can almost hear the Pundit on TV before the second runs of the two winners saying ".. Well he started at 100/1 [or 66/1] FTO so they cannot think much of him..." before being proved clankingly wrong with their hopes that others and the Market will do the analysis for them. Here are the 2010 figures :-
More evidence to add to the hope that if you are prepared to stick at it long enough the 'Second String' effect will successfully deliver good horses, or bad races, or both in the right combinations. The 11/1 winner was the useful Tazahum on debut for Stoute when Moore ridden. He has gone on as a 3yo to win at Listed level and compete in Group races. The first string was Mujrayaat on his third run and placed again. He has proved an OR90s handicapper at 3yo. Oh, and another 23 juveniles for a major owner producing just 3 winners during the season.
The 'Second String' effect has perhaps reached a pinnacle in the last two years with the developments in the Godolphin training set-up. In 2010 Mahmood Zarooni began training in Newmarket at a separate stables from the long time trainer Saeed bin Suroor. Although united under the same ownership and top level management these really are different stables with different work riders. On the course Dettori will often only ride the Zarooni juveniles with known, or proven, higher class ones and shows loyalty to bin Suroor in many instances with a mixture of multiple runners. Given that Dettori does not work every day nor travel that far the Godolphin runners in maidens can end up being ridden by a wide range of jockeys. Ajtebi, the Zarooni stable apprentice, many senior jockeys including Fallon, Spencer & Hanagan and so on.
This arrangement where there are two strong, but separate, stables and a varying cast of jockeys and the relative ranking less than obvious seems ideal to assist the 'Second String' Effect. It is also worth noting that the Market seems to underestimate the Godolphin maiden runners overall, surprising as that might seem. Zarooni has produced profits backing all his debut and second time runners since he started training. Less clear cut with Suroor but a minor profit. Ignore the early season runners while the stables are cranking into full gear and the profits improve. The table below summarise the 2010-11 returns for all of the Godolphin maiden runners in races with multiple representatives.
| Season | Number of Multiple Races |
Num 1st String did Best |
1st String Wins, SR & P&L |
Num 2nd+ String did Best |
2nd String Wins, SR & P&L |
| 2010 | 32 | 21 | 10, 31.3%, -0.92 | 11 | 6, 16.2%, +23.00 |
| 2011 | 60 | 36 | 16, 26.7%, +20.62 | 24 | 8, 11.0%, +47.50 |
Not a lot needs to be added to the numbers here. Back everything in races where they had multiple runners and you would be doing very well. Higher percentage of the profits from the Second, and higher, Strings but more evidence that the Market is not tuned in to how good their Strike Rate is overall leaving aside Stringy matters. In 2011 the first seven races with multiple Godolphin runners produced Second String winners at 16/1 (twice) and 14/1. All seven races only had Zarooni representatives in so the problem for the Market was not with lining up the relative performance of two different yards.
In short, the Godolphin returns seem pretty strong evidence for a 'Second String' effect. It will be interesting to see how this develops in 2012 with Zarooni's approach and set-up becoming better integrated within Godolphin and more well known by outsiders. Looking at the 2011 races it seemed clear at times that there was some small scale recognition of the value in the Second String Godolphin runners. There were regular examples of the double figure prices being clipped. Not the "..they're Backing this one... GAMBLE.." that desperate Pundits spend, apparently, 24 hours a day fantasising about on TV but longer term thinking punters knowing a value price.
This brief survey found that backing every 'Second String' runner for trainers Fahey & Hannon and owners Godolphin & Hamdan Al Maktoum in 2010-11 would have produced a solid overall profit for all four sets. These profits have been achieved on Strike Rates well below the average for those studied in most cases. A reasonable conclusion would appear to be that the Starting Price Market underestimates the competitiveness of 'Second Strings' overall.
Although more speculative a major reason for this finding could well be that the Market formers, including Pundits of various forms, presume too strong a correlation between the 'String' assignment and the horse's relative competitiveness. In maiden and auction races containing many lightly raced, or unraced, participants the Pundits will not even have their set of unproven tools to work with so overvalue what they believe to be indicators from external sources. This process can develop unchecked in a form of 'Feedback' with no solid markers to control how this sort of 'information' flows among people with no worthwhile insights but wanting to present themselves as 'knowledgeable'.
For a variety of reasons the trainers, owners and jockeys involved may have only a limited feel for how well the various strings will run. This uncertainty will regularly not be reflected in the Market which will assume 'First String' as meaning the best, and competitive, representative and 'Second String' as an almost certain non-winner. As hinted at above trainers and jockeys may be working with big strings across a range of locations and the 'multiple' horses may never have worked together and never have been ridden by the jockeys whose presence on them is supposed to confer their preferred status. Even if the horses have worked together the jockeys involved may have different agendas in the race. Some version of the 'easy intro' or 'teaching them to settle as a priority' as examples. On top of this the competitiveness in a race will depend upon the abilities of the opposition from other stables and this will almost never be directly factored in.
As a final aside it is worth noting that a number of ownership syndicates are clearly aware of the 'Second String' effect and know that it can be used to affect the SP. If you are a syndicate manager and want to ensure the best prices are available for the horse you know to be the best prospect for the day then putting the second colours on it will help to bring about the desired outcome.