Exporting of data & pace lines

[Mitch44]

I received these questions through a PM and thought it may be of interest to others:

Questions 5/25/20


I hope you guys don't mind be asking a bunch of questions. Here goes.

Do you use the export. I have started. No problems exporting data. If so, how do you evaluate all that information. The key info is scattered all over the spreadsheet.

I only export races I handicapped. Should I export races I don't handicap?

Do you use the default pacelines when handicapping a race? It appears the software almost always selects a paceline with highest adjusted speed rating and/or highest POH for like races. Is that good?

The good part of the program making the paceline selections is that the process is the same on every race. My selection process is variable based on what I see. Would it just be better to accept the RDSS selection.

I have so many questions. I'm thinking a lot better than I did before Saratoga. At least I think I am.

Hope you don't mind this onslaught of questions. It helps to hear what the experts think.


Hi . . . ,
I don't use the export at all and couldn't answer any questions about it. Tim doesn't either. We both or keep data by the old pen & paper. I think Richie does as he keeps a lot of data.


No don't export races you don't handicap. Now some will disagree with that. The reason to do only the races you handicap is that it is a reflection of your contenders and picking of pace lines. It will identify a weakness in those areas. Exporting races you didn't hcp. will just cloud the picture whereas only races you hcp. will provide better and precise information. Not only on contenders and pace lines but the type of races you win or lose etc. etc.


As far as paceline selection I verify the line the RDSS program picks and change it if I disagree. The program will only consider the last 3 lines regardless of distance or surface. The program is probably about 90% efficient but does not employ the Sartin guidelines of best of the last 3 at a comparable distance (within 1 furlong) and a comparable surface.
Horses do have a preferred surface and distance. While they may get some Place and Shows they fail to win unless at their best dis. and surface. Also his guidelines produce better prices because most handicappers are fixated or obsessed with the last race or line.


I strongly suggest you use the Preceptor for final selection within the Sartin guidelines. It considers the POR , adjusted variant and everything. Its absolutely the best pace line selector Sartin invented. It correct about 98% of the time. The exceptions come from distance and how fast a horse may decelerate etc. But for you I would not worry about that and let the Preceptor do the heavy lifting.


Even doing nothing to the program one should be able to obtain 50% winners with a two horse betting procedure. Playing the right races and proper betting or money management should provide you with a profit.


Mitch44

[Lt1]

I agree with Mitch. I only enter races I capped into my models. I don't enter mdn or 2 yr old races into my data and note races for 3 yr olds only. Like Mitch I'm also a pen and paper guy.
Tim

[gandalf380]

i too will also only export races that are non maiden and non 2YO. I am not sure how you would export a race you did not handicap. I will often handicap an entire card of races even if I do it after the race but before I know the outcome. I have a set routine and thus I tend to handicap every race the same. I will filter the spreadsheet by distance and/or track, depending on how I have the base spreadsheet set up. Using filters allows me to keep all the races from one track together without worrying about the distance. I am working up a video on how I export and keep my spreadsheets. It will also show how I use the filters.

[Ted Craven]

The program is probably about 90% efficient but does not employ the Sartin guidelines of best of the last 3 at a comparable distance (within 1 furlong) and a comparable surface.

People should make their own decisions about how to select pacelines (including deciding whether to use the lines selected out-of-the-box by RDSS). The default paceline selection formula is the same one employed in Validator, Doc's last work (the PI Paceline Indicator formula). It does not employ the 'within 1 furlong' condition (though it DOES use a comparable surface if any), rather it considers all sprint distances versus all route distances (route >= 8 furlongs) to today's distance. I nuanced the formula a bit, but retained the broad-strokes.

That's why people must think - and choose - for themselves. Use a 7.5f line in an 8f race, rather than a 9 or 10f line (which the program would select for you). Use an 8f line in a 7f race rather than a 5.5f line (which the program would select for you). Use a good dirt line on today's Polytrack surface rather than a poor Polytrack line (especially if an excuse, or a demonstrated previous affinity). Compare the impact of using a few alternate lines, including a 4th back or further if they seem like they may better represent the horse's ability today - in your judgement. Be wary about using a by-far best ever Adjusted SR line. Etc, etc.

Maybe one day the software will permit you to configure all these nuances to automatically select lines they way you would manually. Meanwhile, 'software between the ears' ...

Perhaps RDSS does not use the Sartin guideline of 'within 1 furlong'. Instead, it uses the other Sartin guideline 'same distance structure'. Keep looking in the Follow Ups or reading various members' posts and you'll find LOTSA other guidelines too ;). A WINNING method is ALWAYS a personal one, indeed informed by previous research and practice by our mentors and other smart people, but their enjoyment of their meal will not fill your own belly. Nor will they reimburse your losing bets. All winning, like enjoyment of life, is personal: that's why you have the freedom to choose. Go out, do the work, and forge your OWN method.

Ted

P.S. BTW, if you look in older Sartin software for Perceptor, you won't find it (except in Speculator). Perceptor is post-Sartin, though approved by Sartin (I just didn't tell him the seeds of it came from Guy Wadsworth, his ex-programmer who he just didn't get along with. I evolved it from there). Mitch is right - it is SUCH a useful concept!

[Mitch44]

Ted; " All winning, like enjoyment of life, is personal: that's why you have the freedom to choose. Go out, do the work, and forge your OWN method." I agree with this 100%. But its like Sartin said when it comes to Jazz, if you have to ask what it is or how to play than you don't know what it is. Jazz is improvisation from many years of experience etc.


There are many nuances to picking pace lines that the program won' t get right. No one will ever succeed without receiving a basic course, a starting point or jumping off point. The starting point with Sartin was his guidelines and with experience the user is able to fine tune it like a mechanic doing a tune up on a car. You can call this basic, guidelines or whatever but its a starting point.


In RDSS the starting point is the default of strictly using the last 3 which is much better than any newcomer could do on their own. They should be given some guidelines to get started with the program and that starts their journey.


Sartin 's guidelines takes the user beyond the basics to better prices and a more predicative pace line. His guidelines for using lines is akin to an advance procedure. I totally disagree of using structure because there are too many elements to it For Ex. using a 5F line for a 7 F race. Well it with in the distance structure but it'll cost you many more winners than it'll produce. Ditto with routes which range from 8f to now a days to 12 F. A huge difference. When you go beyond 1 F or over 9F your getting into uncharted waters which include a different pace scenario which requires more stamina class etc.


While I follow the guideline of within 1 F I attempt to stay with 1/2 F when that is possible . Many times its not possible but that my personal guideline because of the deceleration factor of horses.

As Ted say all this is personnel choice and what the individual user chooses to do. In the above answer I didn't try to confuse the questioner with how I do it but to give him a starting point for success. All I can say is that anyone who thinks they can ignore many years of research and Sartin guidelines, I wish them much luck as they'll need it.


Mitch44

[rdiam]

I will probably get "flamed" for this post, but here goes anyway:

STOP worrying about pace lines and START worrying about bet structure: which horses do you want to bet against, how will you handle favorites going off 3/2 or below, are you betting the race vertical (win, place, show, ex, tri, super), horizontal (DD, p3, p4, p5, p6), or both? Are you running separate bankrolls for each type of bet? What percentage of bankroll are you betting? What are your minimum odds requirements for 1, 2 or even 3 horse win bets?

You can even pick your contenders using the top 4 or 5 Morning Line selections (see Yellow Manual) and leave the computer behind!

Stress over the important stuff: how you bet your money.

Richard

[Mitch44]

I agree Richard whole heartily. That's the weakness of most handicapper. I generally take my top contenders and design a bet to capitalize on my analysis and the mistakes in the odds the public is giving me. It works very well for me.


I'm always looking for a lot from a little outlay. You won't get "flamed" by me.Your absolutely on point.


Mitch44

[gandalf380]

With all due respect, I believe that Mitch and Richard, because they are both such good handicappers, are putting the cart before the horse. They already are picking the right pacelines which are producing the correct contenders. For a newcomer or a recent returnee, it can be quite a daunting task to not only choose the right contenders but be confident that they are choosing wisely. This is where exporting data and pacelines can be so important to the newcomer. You can pinpoint what type of races are your weakpoints and which are your strong points. Many years ago, I was absolutely horrible with turf racing. But by studying my paceline selection and studying "Pace Makes The Race", turf racing is now one of my strong points.

Once I realized that the programs are excellent at choosing pacelines, then I focused more on the betting angle. Proper bet structure is not a miracle worker when you are structuring the wrong contenders.

[Old Arkie Gal]

I do use the spreadsheet. I would suggest as gandolf380 has suggested learn how to use XCEL's filter options. Then you can look at any variable that you want to focus on, distance, surface, class, or F&M.
I have had several conversations with Mitch and gone over several races with him and what he has taught me re paceline selection has been invaluable.
Pat

[Auraofglory]

Most of the above discussion relates to "modeling", for which the export function is obviously the way to go.


What about track "profiles"? Can you also access that info from Trackmaster when using RDSS?


Thanks for reading this............Jack

[Mitch44]

FWIW I don't keep a track model and largely consider it a waste a time. Its really over rated. When it was highly rated it was effective. What changed? Well the program changed and became much more refined.

For those that don't want to do the work you can consult the summary of the PP's from Bris. The stats there are decent. Notice I said decent and not great etc. Their downfall is that they use the visual ESP which isn't that accurate as the % Med. within the program. If any one does use the exporter within the program than you should be analyzing % Med by track , sex classes etc.

Well back to the program, the program contains the Match Up of the horses within it . And its the Match Up that supersedes any track profile or model. Also within the program is both the visual and % Med. ESP's. Now this confuses most and far too many default to the visual ESP. If it were my program, which it is not I would totally eliminate the visual ESP as its worthless today.

Few review the horses PP's within the program, there are all the ESP's for every race the horses has ran recently up to 10. Its easy to find the horses normal running style and also a change in running style which can produce some nice prices. This information greatly ups one's game. There are varying degrees of Early for an example with some too fast to win and can be eliminated, while others can adapt to todays POR save energy for the stretch and win, or be an EP or P. These correct ESP's also can help to pick correct pace lines.

I talked about this in my posts on "From the Top" Rather than taking the wrong fork in the road or not looking at the road signs along the way your answer to modeling, track profiles and the Match Up are within the program itself.

Mitch44

[Auraofglory]

Mitch44,


Thank you for your reply. It was clear, constructive and very helpful.



May the monkey continue to thrive.


Jack

[Lt1]

Good post Mitch. I do keep a track model for each track I play by surface and dist. Being a 1 track a day player I find it's still effective for identifying the factors which are dominating at the track at the time. Also %med does help downgrade horses and break ties. That's the beauty of the Methodology each of us is free to use the programs as we see fit. No rules just guidelines.
Tim

[Dorianmode]

I agree with Mitch. I only enter races I capped into my models. I don't enter mdn or 2 yr old races into my data and note races for 3 yr olds only. Like Mitch I'm also a pen and paper guy.
Tim

This is an interesting question, and I agree with this and with what Mitch said. I'm also a pen-and paper guy, but it's altered in a way, as there are too many things that need to be looked at, and the only way for that to happen, for me, is to let Excel do all the numbers, and point me to what it calculates, and to what I have told it to alert me to, when the numbers are exceptional.

The thing I don't do, is enter all races, even the ones I "do" into my data. Some few races defy evaluation, and/or an "outlier" wins. I mark them as such. I want to know how many races of each of the types I look at, "defy" my evaluation ... ie where a totally unexpected horse comes out of nowhere, and wins. It's fairly rare, but I need to deal with that in my break-even eval. It's a way to "write off" a race I missed, re-looked at, and not finding something I should noticed, just move on.

So on each of my race forms, for races I've actually evaluated and bet on, I have a box that asks "enter data ?" It is a big topic in statistics, ... this ... how to deal with "outliers". I leave them out of the data OR at least have a way of being able to select the outliers, to filter them later, as they really do nothing but "cloud up" the picture for the vast majority of races that are not outliers. Keeping a column for "outlier", is easy to put in an Excel record.

You could even make a case for only entering the races which lead YOU to win your bets.

[Mitch44]

Outliers are just what you said and that is their unpredictable. Being unpredictable its doubtful at best that any concrete data can be mined from it.


I've seen races that totally define any explanation. It one thing to a contender go off form and quite another to have 4 or 5 and the race run completely contrary to any explanation. Or in other words the worst horses in the race come into the top 3 or 4 payoff positions. But it doses happen. If it was that simple we all would be better with longshots.


While these occurrences are unpredictable I believe they can be accounted for in one's betting approach.


Anyway good luck with your data search and approach.


Mitch44