Horse racing and betting...
...have been closely associated for over two centuries.  The HOOF  race ratings service publishes accurate, innovative, and informative race ratings to guide your betting, with the emphasis placed firmly on finding good value. We have been on the Internet since the 1995-96 NH season, initially as a free service, and latterly by subscription.  The success of the race ratings and the integrity of the service means that we have stood our ground with a small but loyal nucleus of satisfied users.

The software which generates the ratings uses Bayes' theorem to offer an effective and mathematically principled calculation of every horse's chance in a given race. Embedded in the program are around 2500 Bayesian likelihood ratios, derived from independent factors in a sample consisting of  over 25,000 races and a quarter of a million individual runs.

The principles behind the program were established twenty five years ago by Tony Drapkin and Richard Forsyth in their influential book 'The Punter's Revenge'.  The book is now outdated in some areas, but the statistical principles outlined in the chapters on horse racing and football betting remain valid.  An online version of the book is now available free at the HOOF ratings forum.

Race Ratings
The best of the race rating services currently available in the UK such as Timeform operate on the basis of a universal handicap, in which every horse's ability is expressed numerically, usually in pounds weight. Typically, the basic rating is adjusted to bring it in line with the weights allotted according to the conditions of the race in which the horse is entered. This gives the user a means of rapidly assessing a particular race.

Thus a horse with an adjusted rating of, say, 145 is 3 pounds "better in" than one with a rating of 142. This in turn means that in theory, the horse rated 145 should finish about a length in front of a horse rated 142, depending on the distance of the race.

The words "in theory" are crucial here, because if handicap ratings always worked out, bookmakers would cease to exist, as would betting because the best horse on form would always win. In practice, life is not so simple. In non-handicap races, only around 35% of top-rated horses win, while in handicaps the figure is around 20%, depending on the ratings employed.

A further limitation on the usefulness of handicap ratings is that they are difficult to interpret. For example:

• How much more confidence can the punter have in a horse rated 5 pounds clear compared with one rated 2 pounds clear?
• Is it better to restrict one's betting to non-handicap races where the horses with an obvious chance stand out?
• How much should one stake on a horse which is clear top-rated?
• What odds should one be looking for ?

Value Betting
In any race, there is no such thing as a certain winner unless the race in question is a walk-over. The so-called "good thing" may get beaten, even at long odds-on in a two-horse race.  It follows that one is never engaged in picking the winner of a race, merely the probable winner.

The scale of probabilities ranges from 0 (no chance at all), to 1 (absolute certainty). It is relatively simple to convert probabilities into odds against using the following formula:

Odds against = (1 / Probability) - 1

The essence of value betting involves backing only those horses whose odds are longer than they should be. Many professionals use a version of this strategy, since it is the only one which offers any hope of long-term profitability.

Thus a horse which is a true 2/1 shot is good value if offered at 11/4 or better. If one can consistently find such bets, one should, in theory, win in the long run. Even so, this is no short cut to easy riches. Finding value is hard work, and any backer who can make 10% on turnover is doing very well indeed.

HOOF value betting ratings are designed to make the business of spotting value considerably easier.

Using HOOF Race Ratings to Find Value Bets
The output for a typical race looks like this:

Ratings for 30th April 2007

                      Towcester:3.40
                     Handicap chase:9 rnrs 
HORSE                TR-PROB TR-ODDS   V-30  V-PLUS
Missy Moscow(IRE)      0.223    7-2    11-2    7-1

Metal Detector(IRE)    0.19     9-2     7-1    9-1

Earcomesannie(IRE)     0.139   13-2    10-1   14-1

Alf's Spinney          0.098   10-1    14-1   20-1 Non-runner
Colonel Sparks(IRE)    0.092   10-1    16-1   20-1
Cockspur(IRE)          0.08    12-1    18-1   25-1
Monsieur Georges(FR)   0.075   14-1    20-1   28-1
Auditor                0.058   18-1    25-1   40-1
Magic Of Sydney(IRE)   0.042   25-1    33-1   50-1

The top three are clearly indicated.  Where there are joint top, second-top or third-top rated, more than three horses may figure in the top three.

See my  criteria for ranking the top three.

• The TR-PROB column gives the 'true probability' of the horse winning, as calculated by the program using Bayes' theorem on a database of 25000 races.
  
• The TR-ODDS column translates the true probability into the corresponding odds.
• The V-30 column gives the minimum odds at which a horse may be considered a value bet, with a 30% margin in the backer's favour.
• The V-PLUS column indicates outstanding value bets with a 40-45% margin in the backer's favour.

Most of the value bets to be found in the top two rated start at an average price of around 9/1. This necessarily entails a fairly low strike rate of around 10.5%. Coping with the inevitable longer losing runs is simply a matter of sensible staking with a larger bank, typically 200 points, compared with the 20 point bank which is recommended in many of the standard texts on the subject.
 

Betting Strategies
The ratings do not in themselves constitute a "fire and forget" tipping service.

The user must be selective and employ a certain amount of judgement. Among the factors not built in to the ratings are allowances for trainer, jockey, class, and going.

 No horse should ever be backed unless its odds are equal to or greater than its V-30 odds. At the very least, the ratings will save the user wasted bets by clearly indicating false priced favourites.

 Any horse whose odds are equal to or greater than the V-30 odds, or better still the V-plus odds, particularly if it is  in the top two or three rated is worth serious consideration.

There may well be two or more value candidates in a particular race.  Recent experience on the betting exchanges (see below) suggests that it is now perfectly possible to back two or more in a race, provided that the price is right.  As always, the prime aim of the program is to enable selective, intelligent, value betting.

Getting Best Prices

1.  Early Prices and the online market
One of the advantages of playing in handicaps is that in many cases, one can bet at early morning prices offered by the major bookmakers and many of the independents. These early prices are made up by expert odds compilers, and put out by the bookmakers in order to generate interest in handicap races, where in theory their profits are likely to be greatest.

However, the early price market is worth serious consideration for a number of reasons:

• The backer is pitted against individual odds compilers, rather than the market as a whole.

• Odds compilers are not infallible. They can, and frequently do make mistakes. It is not uncommon for a horse to be priced up at 8-1 in the morning and to run out a winner at three or four points shorter in the afternoon.

• Different compilers will have different opinions, which will be reflected in the odds quoted. If one concentrates on the best price available for each horse in a race, it is frequently possible to bet against a collective overround which is much smaller than would normally be the case.

The online market  offers many opportunities for value-seeking before the start of racing.  In addition to the many bookmakers who offer early odds, there are the person to person betting exchanges such as Betfair and Betdaq, where punters may either back or lay horses from early morning right up to the off , and even in running (see below).
 

2.  Taking advantage of on-course market movements
Backers who can follow market movements, either in a betting shop or at the course, and take advantage of best prices, will find the ratings profitable.  Many shrewd backers like to take a price "on the drop", i.e. the first price after a horse's odds have shortened.  For example, if a horse is quoted successively at 11/4, 3/1, 11/4, 5/2 then one takes the second quote of 11/4.  If a horse opens at 5/2 and shortens on the next show to 9/4, then 9/4 is taken.  This technique does not always guarantee the best price, but it generally ensures a price better than sp.

A technique some shrewd backers use is to work out the overround on each show.  Allowing 2% per runner as a guide to the probable overround,  in a field of eight runners, the bookmaker's take will generally be  around 16%.  If, in the course of the SIS shows the overround is favourable (e.g. less than 16% in the example quoted), then that is the moment to back your chosen horse at its current price on the grounds that you are unlikely to better those odds. On average, this is around the third or fourth show.  In this way it is frequently possible to get what turns out to be the best show.

The dedicated skirmisher with a nose for the market, able to think on his feet  and get that all important best price, has a far better chance of coming out on top than the average betting shop punter, or the simple racegoer out for a bit of fun.

3.  Person to person betting exchanges
Person to person betting sites  such as Betfair have been around on the Internet for several years now, and are becoming  increasingly popular with backers.  They are deservedly attractive  because they offer enormous flexibility, in particular the opportunity to place odds-conditional bets.   Anyone who is prepared to take a view about the chance of a horse in a particular race, can choose to act either as a layer or a backer of the horse in question.  Bets from those wishing to back a horse at a particular price are matched against odds offered by those wishing to lay  it.  Those who run these sites claim  that  it is frequently possible for backers to better sp.  That has certainly been my experience in recent seasons.

Another advantage of betting exchanges is that of laying, as an alternative to backing (in effect, betting that the horse will lose).   Although I have no experience in this area, being a backer by temperament, I am told by some subscribers that the ratings are a useful tool in assessing short-priced favourites that are worth opposing.

Wherever you choose to bet, you should always make a point of playing the market and getting the best price you can.  As in futures trading, this can make all the difference between success and failure.
 

Losing Runs
Losing and losing runs are not a popular topic among most services, many of which prefer to design their advertising copy so as to suggest that making a money from betting is easy.  If you do not know so already, you will soon discover that it is not, and that those who suggest otherwise are doing their subscribers no favours.

With  a 50% strike rate, the chances of a losing run of 20 are over a million to one against.  However, with a 30% strike rate, this figure comes down to just over 1000/1 against.  With a 20% strike rate the chances of a losing run of 20 are a mere 85/1 against.

Whatever your betting strategy using HOOF ratings, it is highly unlikely that you will be backing favourites, and more than likely that you will be betting 'against the crowd' i.e. in an inevitably high-risk area in which long losing runs are a fact of life. My worst losing run last season was in the high 40s.

All HOOF subscribers are urged to take to heart the suggestions laid out in the section on Betting and Common Sense below.

Betting and Common Sense
Betting is a high risk activity and making a profit is never easy.  Value betting as advocated on this site offers no guarantee of instant riches. Finding value generally means not betting on favourites.   It is inevitable that for those choosing to bet on outsiders, there will be long losing runs when nothing seems to go right.

In all honesty, I have to record the fact that I have had some dreadful losing spells over the years, and that I have been saved from total disaster by three things:

1.  A 200-point betting bank.

2.  Small stakes.

3.  The stop-loss option.  In the event of half my bank being lost, I am committed to halving my stake.

4.  The keeping of scrupulous Excel-based records.  I am always aware of exactly how much I am winning or losing during the course of a season.

I publish 'warts and all' details of betting strategies in spreadsheet form on the HOOF discussion forum.
 

 It is therefore best to put by a bank of at least 200 points and to bet with modest stakes.

Those who advocate a starting bank of 20 points, even with a high strike rate (and very few punters can manage a strike rate much better than one winning bet in three), are completely ignorant of both the psychology of betting and the likelihood of long losing runs.

 The unpleasant truth is that even with a successful strategy and a high strike rate, losing runs in the high teens and twenties are quite common. Consider now the pressure on the backer whose starting bank is 20 points following a losing run of fifteen or more. At this stage, confidence begins to ebb, and the decision to bet or not to bet becomes fraught with conflicting emotions.

 A backer who has a starting bank of 200 points, and who is betting with money he or she can afford to lose, may not relish a losing run in the high or twenties or thirties, but will be in far better shape to cope psychologically than the backer with a much smaller bank.

My advice to all backers, whether they be HOOF subscribers or not,  is to bet exclusively with money that can be lost with equanimity, and to divide the starting bank into 100-200 units, depending on anticipated risk exposure, stakes, and turnover.

Always remember that you are not compelled to bet, even if a horse on the strength of the ratings appears to represent outstanding value. Check out the form, particularly those factors which are not covered by the HOOF Ratings program, and especially the going.
 

Withdrawn Horses
Last season, as in previous seasons I was struck not only by the large number of non-runners, but more importantly, by the unacceptable frequency with which the notification of withdrawals is delayed until well after the start of racing.

My research indicates that around 35% of all races will have a horse or horses withdrawn at any time from publication of the overnight declarations to just before the off.  In a small number of  instances, the withdrawal will be indicated in the morning's racing press under the rubric, "Trainer states that X is a doubtful runner."  Many more withdrawals are made around mid-morning of the intended runner's race, and communicated to the racing public on SIS screens in betting shops or via the Racing Post or Sporting Life websites.  However, a significant number of withdrawals are not made public until well after racing has commenced.

This should give cause for concern for the simple reason that the shape of a race can be transformed by late withdrawals.  Of course this is not always so.  In some cases, where a non-runner's true probability of winning is less than, say, 0.0476 (equivalent to a 20/1 shot), the effect of withdrawal on the shape of the race and on the betting exchanges, is likely to be minimal.  However, in races where a horse with a reasonable chance is withdrawn, or where the cumulative effect of two or three withdrawals amounts to a probability greater than 0.0476, it is inevitable that the chances of the remaining horses will have to be recalculated by the discerning, value-conscious punter.

 Around 25% of all races are affected in this way, and that in many cases the betting public are given insufficient time in which to reach an informed decision on what to back in the new version of the race, or indeed whether to bet at all.

Please note that when horses are withdrawn or become non-runners, the ratings will be affected.  The effect will be negligible if the withrawn horse is a rank outsider, but considerable if it has an obvious chance on form.

Where updating is impossible, there is a simple adjustment formula which allows anyone with a copy of the ratings and a pocket calculator to revise the ratings in the event of non-runners. All subscribers are urged to print out the adjustment formula and to keep a copy handy on days when they are proposing to bet.

Alternatively, all subscribers are given an Excel-based utility to enable them to recalculate the ratings in the event of non-runners.