More Betting Thoughts: What Trends To Look For In A Horse Racing System…

The cold continues, dear reader, ensuring my nose and throat are as full as my inbox currently. And it's directly to the inbox that we'll go, as promised yesterday, to answer publicly another reader's question.

Steve, from Tasmania no less (you devil!), wrote in to ask about systems research, specifically in relation to trainer patterns. His email is below, and my answer then follows:

From: Steve W
Sent: 18 December 2008 23:27
To: info@ATgeegeez.co.uk
Subject: Geegeez Question

Hello there Matt,

Firstly - I hope you're feeling a bit better?? Nothing worse than being crook!

My name's Steve, and I'm in Tasmania - Australia.

I've been looking up your stuff and read some of your articles - excellent stuff!

To be brief:- I was hopeless in my 20's on the punt. I read some good stuff (good books), but had neither the capital nor the discipline to bet sensibly. I had a good long break from punting that did me the world of good.

Now, a bit older (33) and wiser(!) I'd like to make a fresh start. One of the things that always intrigued me was trainer stats. I have a couple questions, if you don't mind. I'm hoping you might swing me in the right direction with a bit of advice and guidance?

(a) Have you found any books particularly helpful? Been a long time since I've bought any.

(b) I intend to purchase a database of some sort for Austalian racing, so I should be able to have a look at some systems. I'm looking forward to that. Any advice there (other than, keep the rules logical?) I note with interest that you don't mind being track specific in your mechanical systems that you have posted in some of your old newsletters? Have you ever gone back to see how they've held up? Can a portfolio of systems approach win overall?

(c) Trainer stats! Where do I start?? lol Here is where I'm really hoping you might help me on my path. Obviously there's a lot of things one can look for. Are you able to guide me in what you've found to hold up over time? When I've had a cursory look, I've noted that stats between different trainers in broad categories just don't seem to hold up year to year (e.g. one year J Smith made a profit with first uppers, the next year you would have lost following them). In fact, so far, (albeit from a cursory look), the stat that seems to hold up strongly is that some trainers are very much overbet on there short priced horses. So - either lay them, or, the overbetting of them creates value in the race.

In the DRF form guide (for the USA) I noticed that they track the trainer patterns (switches) - things like blinkers on, big class changes etc. Is that the sort of thing you are looking for??

One thing I've noticed here in Australia, is that a trainer who has a really good strike rate at that particular track, do seem to win a good few races!!

So - I'm looking for advice really!

What type of stats to track

How to use those to bet! (i.e. does one still handicap the horse etc)

Finally - I'm interested in your answer if I were to ask you: "why do you fundamentally believe that your trainer stats method will continue to be an underbet (therefore profitable) method into the future?

If you've read this far - many thanks for your time! It's a big ask of a stranger, but I'm really hoping you might help me out here!

Blessings & merry christmas to you,

Steve

--------------

Hi Steve

Many thanks for your long email, and sorry for the delay in coming back to you. I’ve been under something of a deluge just recently! Still feeling rough, I’m sorry to report on the sickness front, which of course doesn’t help anything.

Regarding your questions

(a) I’m not sure if you’re referring to Australian racing (about which I have no knowledge!), or UK racing. If the latter, there are some excellent statistical compilations, such as Raceform’s ‘Trainer Jump Statistics 2007-8’ (ISBN 978-1-905153-78-7), in which you may find some lucrative lesser known trainers with profitable sub-trends.

There is also a flat racing version of this book, called ‘Trainers Flat Statistics’, ISBN (978-1905153749).

Both of these are available from Amazon, and all good sports book resellers.

I do of course produce my own jumps statistical guide, called ‘TrainerTrackStats’, which you can get from www.trainertrackstats.com. I’ve produced this in conjunction with Gavin Priestley this season, and it continues to show a profit as it has done in every previous season we’ve produced it.

(b) and (c) Regarding what sorts of parameters to look for when researching a system, if the system is based on trainer patterns, then it logically follows that you should consider possible differences in the way trainers train. Taking that a step further, I mean that here in the UK, some trainers are more effective with horses run over shorter trips or on sharper, i.e. less stamina sapping courses. Some perform best when there is a real slog.

For instance, Nigel Twiston-Davies has the following record over the last four seasons (14/1 odds or shorter):

2m races 68 wins from 332 runs (20.48% strike rate) Loss of 34.62 points

2m1f – 3m 145 wins from 781 runs (18.57% strike rate) Loss of 40.22 points

Your first 30 days for just £1

Over 3m 48 wins from 236 runs (20.34% strike rate) Profit of 56.58 points

Although the strike rate remains consistent, the profit emerges only in races with more of an emphasis on endurance. (Incidentally, this implies that his horses are overbet at shorter distances, and underbet at longer distances.

Taking another example, let’s look at Alan King’s record over the same period:

2m races 128 wins from 571 runs (22.42% strike rate) Loss of 62.31 points

2m1f – 3m 184 wins from 870 runs (21.15% strike rate) Profit of 47.27 points

Over 3m 30 wins from 146 runs (20.55% strike rate) Profit of 22.31 points

Although on the face of it, it appears that we should back King’s runners at anything beyond the minimum trip, slightly deeper research reveals that for trips beyond 3m, you would have lost in three of the four seasons in questions, and a couple of big priced winners in 2006/7 season have skewed the stats somewhat.

Trainers may also train horses for chases rather than fences, or vice versa (or even for bumper / flat races).
Consider the following over the last four seasons:

Alan Swinbank (3 and 4 year old horses)

Hurdles 3 wins from 17 runs 17.65% strike rate

Bumpers 27 wins from 75 runs 36% strike rate

Nicky Henderson (all horses)

Chases 73 wins from 358 runs 20.38% strike rate

Hurdles 159 wins from 659 runs 24.13% strike rate

Bumpers 51 wins from 159 runs 32.08% strike rate

Some trainers focus more on precocious early types (both the above are examples of this), whereas others tend to allow their horses more time to mature.

Some trainers focus on setting horses up for nice priced handicap wins, so the race type is another area to look at.

Some trainers see their horses perform on certain ground conditions (often, though not always, allied to whether they train horses for stamina or speed).

And all trainers go through good and bad runs – in some cases, this is ‘deliberate’ in so much as they train their horses en masse for certain key targets, such as the Cheltenham or Aintree festivals.

Although in the above examples I’ve tried to find positive indicators, do bear in mind that with all of these, you may well uncover strong negative patterns that can be used to handicap a race without a certain runner (or, of course, to lay the negative trend horses).

Finally Steve, in answer to your question about why I think TTS will continue to be underbet and therefore profitable to follow, this is simple. We don’t make fortunes and most of my customers are small players whose total cumulative investment measures more as a ripple than a splash in the great liquidity ocean that is Betfair (where virtually all TTS players wager).

In other words, TTS is not mainstream. Even though a few hundred people now use it, that’s as nothing to the tens of thousands who play on Betfair. And, moreover, when the average TTS punter stake is factored in against the average Betfair unit stake, again TTS is not hitting the pools hard. Of course, this may not last forever. But for now, I’m comfortable and confident that things will continue ‘as is’.

Hope this helps, and if you do find anything interesting in your Oz research, perhaps I can help you market your system (if that’s a route you’d like to take).

Best Regards,
Matt

********

I will try to feature some of the more interesting letters on here going forwards, as this seems to be a popular feature in newspapers and magazines (I assume for their reader interest rather than just as filler - I hope you don't consider the above to be the latter!).

********

Onto Project Betfair, and after a no bets day yesterday, today's runners are:

Hereford
2.10 Six Day War

Ok, so that should have been today's runner is... but no matter.

********

Finally, although its only Monday, I heard a great joke yesterday from my old friends Ronnie and Ronnie (Barker and Corbett, lest you're not following), in the 'And here is the news' section of their timeless, peerless Christmas Special of many moons ago:

A woman from Worcester was seen covering her horse with lettuce and tomato yesterday. When asked what she was doing, she replied that she wanted to ride side salad... Boom, boom!

It may just be me, but I thought that was brilliant.

Until tomorrow,
Matt