Cheltenham Festival 2016: Looking for a winner? Ignore the 'trends'

I’m looking for a Capricorn, 13-year-old bay mare, with a white fetlock, who finished 9th in its last race, at 16-1
I’m looking for a Capricorn, 13-year-old bay mare, with a white fetlock, who finished 9th in its last race, at 16-1

Jack Houghton will be using trend-analysis to highlight some value bets at this year's Cheltenham Festival, but before he gets started, he outlines why most of what is written about 'trends' is nonsense and should be treated with suspicion...

"I blame the editors, 64% of whom didn’t pass their Maths O’ Level."

In the past, I've been quite rude about racing journalists who peddle so-called "statistical" trends in their writing.


And I'm going to continue to be as rude as I can here: because in just the same way that it's wrong for actors-dressed-up-as-dentists to recommend a brand of toothpaste, it's also wrong for mathematical simpletons masquerading as Stephen Hawking to be allowed to sell voodoo as if it contains some kind of analytical truth.


I blame the editors, 64% of whom didn't pass their Maths O' Level.


Thankfully, the wonderful editors at Betting.Betfair are not part of this scurrilous bunch, and they have hired a statistical expert - me - to cast my eye over each day's racing during this year's Festival and use what trend analysis I can to point out some value bets.


Before we get to the racing, though, it's worth recapping previously-trod ground outlining why most of what is written about trends is a load of tripe.


First, most of them pick unusual time periods over which to analyse any trend. For example, in the same publication, I've seen a 32-year time-frame used for the Champion Hurdle, a 13-year time-frame for the Queen Mother Champion Chase, and a 17-year time-frame used for the Gold Cup. A closer look soon explains the seemingly random date selection: go back one more year and the "trend" is bucked.


Second, most of them focus only on winners. For example, I read recently that five-year-olds are only "one-for-94 since 1985". A five-year-old won in 1985, of course, but leaving that aside (see above), focusing on winners is statistically irrelevant. For the trend to be of any use, you need to know how many of those five year olds finished in each place, from first to last, and how this compares to horses of different ages. (The answer is that five-year-olds are marginally less successful than seven-year-olds, but it's hard to find a profitable angle from the analysis).


Third, allied to the point above, many of the trends fail to be especially discriminatory. Take this year's Gold Cup. A publication tells me that the winner is likely to have won this season (groundbreaking stuff), which eliminates two of the fourteen runners, both of whom are triple-figure odds.


Fourth, also allied to the point above, many of the trends fail to focus on profitability. It might be true that Champion Chase winners in the past have tended to have previously run in the same calendar year (not on December 31st, though, absolutely not), but backing all those horses who have done likewise does not return a profit in the long-term.


Which is why trends that focus on where winners came from in the SP betting are so ridiculous: is it really useful to know that winners of the World Hurdle tend to come from the first four in the betting? Horses with shorter odds are more likely to win races than horses with longer odds. That's all this trend tells us - something punters hopefully already understand.


There are some trends worth clinging to, though. The most useful is that which says that previous Festival form is an advantage. Simply backing horses who have finished in the first four in a previous Festival race has returned a level-stakes profit at Betfair SP over the last six Festivals (there's no significance to the time-frame here, or the arbitrary choice of first four - it's just as far back as I could be bothered to go and is meant to be illustrative).


A more discerning approach, though, is to use some kind of ratings, such as those provided by Timeform. If a horse has been able to produce a near personal-best rating at a previous Festival, then it has a strong chance of being able to do so again.


The reason for this trend is that racing at the Festival is more competitive, and therefore usually more high-paced, than most other racing, and if a horse has shown themselves able to cope under these conditions in the past, they are more likely to be able to do so in future.


Where previous Festival form is limited in a race, though, it is still possible to utilise the spirit of this "trend". Using speed ratings can help you to identify which horses have shown themselves capable of racing at the break-neck pace of most Festival affairs.


It will be an amalgam of this Festival-form trend - and anything else I come up with along the way - that I'll use to identify the value bets at this year's meeting. In the meantime, ignore the know-nothings who promulgate largely useless trends - they'll harm your profitability more than they'll help it.

Day One Offer: Back a 3/1 winner and get THREE free bets

Yes that's right - back any winner at 3/1 or higher on Tuesday at Cheltenham and we'll give you three free bets to the same value as your stake, up to £/€25. Sportsbook bets only, T&Cs apply.

Day One Offer: Back a 3/1 winner and get THREE free bets

Yes that's right - back any winner at 3/1 or higher on Tuesday at Cheltenham and we'll give you three free bets to the same value as your stake, up to £/€25. Sportsbook bets only, T&Cs apply.

Get a Free £/€20 Exchange Bet

  • Join Now - Open your account today using promo code VAL225
  • Bet - Place a £/€20 Bet on the Exchange
  • Earn - £/€20 Back in cash if your bet loses
Bet Now

T&Cs apply.

Discover the latest articles

Read past articles