Foward To The Past

Foward To The Past

There are always sharp handicappers out there from whom we can learn and make our own game better. 

Many of those sharpies are ‘figures out of the past’ – players who have had their day in the sun and are now either deceased, or whose writings have generally been relegated to the dusty shelves of old handicapper’s (like me) libraries.  

But I have always been one to go back and study those guys – attempting to glean some golden nuggets that will apply to the game as we currently play it.  Though racing is continually evolving, the value of what these giant figures of the game’s past have to offer shouldn’t be overlooked.

Recently I was reminded of the great statistics compilation done by Fred Davis in the early seventies . . .

We used one of his findings in the formulation of the DA Key method (originally released to the restricted membership of our Insider’s Club).  That statistic was: 

Davis found that, in sprints, if you averaged the two best unadjusted speed ratings (ratings as found in the DRF) from races of the same approximate distance (5.5, 6, 6.5 and 7f) – the horse with the highest average won far more races than their odds predicted they would.

In fact, the results were almost shocking . . .

Those highest average horses won 238 per cent of their fair share (according to their final odds) of the races in the study!  He then organized the remaining horses whose two averaged speed ratings fell into specified groups;
– up to 2 pts. below the top,
– between 2.5 and 4.5 points below the top,
– between 5 and 7.5 pts. of the top,  – etc. 

Horses in those first two groups only – that is, up to 4.5 pts. lower than the top – were profitable . . . while horses whose average was in any of the groups starting with 5 or more points lower  – were overall losing bets.

Considering the average number of horses entered in a race (though sadly – that number has fallen precipitously since Davis’ time), a handicapper would be able to cut the number of horses to handicap to only about 1/3 of those entered, but in that group would be found 2/3 of the winners!

Now, using adjusted speed ratings, daily track variants, and various other handicapping factors will give a clearer picture and a better ROI, but even by just using the unadjusted raw speed ratings as did Davis in that study, you would already be placing yourself in the winning ballpark. 

Do some testing of this for yourself – combine with other methods – whatever.  You’ll improve your contender selection greatly in the process.

And a final word – it works for routes as well – but in the Davis study, only for the highest average SR horse (no other ‘groups,’ like mentioned above, offered a profit.

Best of fortune to you, and as always – comments welcomed                      – Gary


  1. OK, this is just preliminary, as it takes a little to get used to doing the figuring, but using BRIS, passing turf races, I’ve hit 4 of 6, for about a $20 return on $12.

    I find it a bit confusing when adjusting for highest ever. Do you credit for highest ever before track/Mdn adjustments? That’s how I did it, anyway. I’m also going to make a new track chart in alphabetical order, rather than order of rank. That would make finding specific tracks easier.

    BTW, do you count turf speed ratings for figuring, if race is dirt or AW? I was going ahead and using them, as the rules said nothing about it.

    I can see potential using these as main picks in Pick-3 and Pick-4 plays. Still need to practice more, tho.

    Thanks for a method that appears to have potential!

    • John – Good to hear of your early success. 1. I pick the “highest ever” from the raw ratings. If you find adjusting first for maiden/track class – then halving the diff from resultant figures – is working – use it! (I’d suggest testing both ways though so you have a results comparison). 2. Comparing turf and dirt or poly is always a ‘touchy’ thing – if the horse has shown no races on the ‘other’ surface, of course your choices are only use what you’ve got, or pass the race. If a horse has shown failures without an apparent excuse on the ‘other’ surface, I’d be wary about using higher rating races from that surface (or indeed the horse itself?!). – Gary

  2. Just got the method. Looks intriguing. I have two related questions. One, you said Beyer numbers not recommended. But the original concept of Fred Davis used the DRF speed figures, without the variant, right? Weren’t the Beyer numbers supposed to be an improvement on those?

    So, since all my pdf files of Past Performances are BRIS, I wonder if those would work. Thanks.

    • John – Beyer ratings use par comparisons as a way to also integrate class into the final rating. I believe that extensive research has shown speed ratings to be as accurate or more so in regards to isolating the potential contenders in a race. As I mentioned – using daily variants could improve over raw ratings – IF – those variants are accurate (which they often times are not) which would pretty much require making your own.

      But Beyer numbers and Bris numbers – whatever numbers could work – as long as you have worked out the comparison ratio. I.e. rather than the profitable cutoff being 5 points below top – if using a different set of ratings that number might be 7, or 4. It would take some research to establish how the numbers compare to the DRF numbers.

      There are couple of posts where we’ve discussed this issue that might be of help – the main one being here:
      – Gary

  3. Now I’ll have to go back and refresh my memory of DA Key. But sometimes I feel like the butt of that joke about the guy who had one sure fire simple method of escape and used it successfully when the predator approached vs. the “smart” guy who had multiple options available (a whole library of systems), but was doomed because he couldn’t decide which would be most effective to use at the critical moment.

    • Dave – Well, you nailed it brother. That is the great challenge in this game . . . sensing when to use certain factors (methods), or when to ignore them in favor of other factors, or when to pass the race altogether. It’s easier if a player limits his/her game to one specific type of race (claimers only, maidens only, allowance only, etc.), but still there are regularly subtle nuances in today’s race mix that may require a slight (but crucial) adjustment in one’s thinking and/or application of his preferred approach. But – hey, it wouldn’t be near as fascinating if it was easy! – Gary

  4. David Stewart

    How far back do you go to find the two best races? 10 races? 3 or 4?

    • David – In the DA Key method we suggested 5 races. Of course – as with any “black-and-white” attempt at a rule in this game – there are times that might not be optimal. How long ago was the race/s in question? Are you gong back to 2 yr old racing if the runner is 3 now, or going back to 3 yr old racing if the runner is 4 now? Were there 4 turf races in the last 5, but today is on the dirt and going back 6 would get another useable dirt rating? Etc. etc. Common sense goes a long way – as does being flexible. – Gary

  5. Gary …

    you are giving the DA away …far to cheap!! (take it off the

    market or make it $47 or better yet $57 )

    I keep going back to it …..simple and last weekend I

    hit 3 for 3 at Woodbine $9, $5.90, $16.40 . Its fast

    and with this new information even more deadly.

    • Joe – Good to hear you are still using the method after all these years. I agree on the price thing, but every now and then I like to offer a method at a near give-away price (or gratis). It seems right. – Gary

      • absolutely ……was (half) kidding …

        but the time savings is a big thing for me.

        I remember thinking it was toooo simple and this

        was one to just set aside (which I did for a couple

        months). One day I started to check some old

        forms and some results, and I keep going back

        to it ….especially on sprints . ANOTHER good

        one Gary …..

Leave a Reply

Your email address will not be published. Required fields are marked *