DO NOT BUY BETFAIR DYNAMITE!

Dynamite, dear reader, is a deadly substance, and a truly graphic word used to 'explode' sales.

If you'll forgive that awful pun, I'll share with you some insights into the latest product offering from the seemingly never-ending Steven Lee Jones conveyor belt.

It seems Mr Jones has got hold of some software similar to that which I use to research TrainerFlatStats (TFS) but, instead of painstakingly number crunch through literally hundreds of trainer patterns at each track, he has decided to run a broad search query to identify some - apparently - profitable trends...

But what Jones has produced is to me a classic example of 'back-fitting' the data. There is no obvious reason to select some courses and not others.

Whereas in TFS, it is obvious that certain trainers target certain races that suit the way they train, and turn a decent profit as a result, with Betfair Dynamite, the 'system' criteria are somewhat arbitrary in my opinion.

For instance, why he's selected the courses he has, rather than all courses, can only be because the results for the other tracks are not so lucrative. In fact, they'd lose money.

The 'system', such as it is, consists of a 26 page guide. TrainerFlatStats runs to 66 pages. Within Betfair Dynamite's 26 flimsy pages, you'll find thirteen pages of the usual filler (how decimal odds work, how to use betfair); the 'system' itself spread over four pages; and the rest containing examples on how to apply the rules. TFS has six pages of 'How To', and sixty pages of content. Four pages of content vs sixty pages of content... Hmm...

Despite the nasty feeling I had about Betfair Dynamite, I thought I'd check out the results using my own database. Now there are a couple of differences between the 'system' criteria stipulated in Betfair Dynamite and the flexibility of my tool (oo-er missus!).

These are as follows, so you can make up your own mind about how material they'd be overall:

1. Betfair Dynamite says, no bet if the horse ran in the last two days. My database can only cater for no bet if the horse ran in the last three days (!)

2. Betfair Dynamite says only back clear favourites. My database cannot split between clear and joint favourites. (This may be slightly material in my opinion)

So how did the results come out. Firstly, I found that the 'system' did actually make a (small) profit over the last two seasons:

YEAR	WINS	RUNS	STRIKE%	LSP
2007	570	1950	29.23	35.10
2008	140	424	33.02	12.26
	710	2374	31.12	47.36

But if you look at the last five seasons, you see the following:

YEAR	WINS	RUNS	STRIKE%	LSP
2003	238	943	25.24	-65.43
2004	273	1086	25.14	-88.25
2005	356	1403	25.37	-61.88
2006	432	1625	26.58	-48.22
2007	570	1950	29.23	35.10
2008	140	424	33.02	12.26
	2009	7431	27.43	-216.42

That's a level stakes LOSS of £21,642 to Jones' recommended £100 stakes. Unsurprisingly, Jones uses the start of the 2007 season for his betting bank calculations. TFS is based on performance over FIVE years.

Two things to say in the interests of fairness at this point:

a) The two criteria that I cannot exactly replicate may have an impact on the bottom line, so Betfair Dynamite may have performed slightly better than my numbers indicate. It may also have performed slightly worse. But I expect that the differential would be minimal.

b) Trends do change over time (as in TFS), so it may be that going forwards the positive pattern is repeated. Maybe.

For me though, I feel that this is a lazy product, all about the marketing over substance. It's something I could have thrown together in an afternoon and, if I had (which I wouldn't because of the dodgy statistical approach), it would be on my 'free systems' page.

[STOP PRESS: JONES IS USING ADRIAN MASSEY'S FREE SERVICE. The results for the 'system' in 2006, the only other year for which there is data, amounted to a loss of £5,317 to £100 stakes.]

If you're tempted to spend $77 on this, then I seriously recommend you spend $20 (about a tenner) more on a properly researched and proven product.

On the subject of TrainerFlatStats, there have been four qualifiers since I last wrote to you:

Oasis Wind was 3rd at 8/11

Wigwam Willie was 3rd at 17/2
Partners In Jazz was 9th at 11/2
Skhilling Spirit was 10th at 8/1

Note that the last named three, bizarrely, all ran in the same race, and were drawn 3, 5 and 8 out of 14. TFS states regarding the draw at Newcastle, "High drawn do best in bigger fields".

The winner was drawn 15 (out of 14, there was one non-runner); the runner up was drawn 14; Wigwam did very well to get third from stall 8.

Partners In Jazz and Skhilling Spirit were drawn 5 and 3 respectively and had little chance of overcoming their positional misfortune.

You've missed very little of the action since the start of the turf season, but things hot up this week from Wednesday, so if you haven't already availed yourself of a copy of TFS, have a look now.

*********

A quick reflection on the Grand National and, despite my pessimism about the stats this year, they were all true again and Comply Or Die (recommended here at 12/1) did us another turn, after Silver Birch at 33/1 last season.

I really do look forward to the race every year, as these trends are fantastically strong and based, unlike Betfair Dynamite, on rock solid logic and decades of evidence.

I'll be back in the next couple of days with the fifth of my six trainers to follow. [Incidentally, you may have noticed that Tom Tate, one of our flat trainers to follow (and an excellent trainer generally in my opinion) won the Aintree Bumper - a flat race for jumpers! - with 14/1 shot Honest John]

Until then...
Matt