If you divided the total by the number of races it could be useful! — Mark

That would give an average SR, but that wasn't the point of the calculation.

Dave's idea is to take the sorted 6th highest figure and look for an 8/1+ horse that is within that 1st to 6th range. I think Steve probably gets it now.

I think Steve probably gets it now.

Yes.

What is it Dave said in one of his workshops "Often wrong, seldom in doubt".

Dave's idea is to take the sorted 6th highest figure and look for an 8/1+ horse that is within that 1st to 6th range. I think Steve probably gets it now. — RanchWest

Not 6th but Top 6.

What is it Dave said in one of his workshops "Often wrong, seldom in doubt". — Steven

One of my favorite quotes, along with one from the TV series, Billions: "I am not uncertain."

Not 6th but Top 6. — Dave Schwartz

I continued with

within that 1st to 6th range — RanchWest

.

Flogging that dead horse. lol

I understand, I was just saying that would be better than how it was. It's like what All-Ways software does with listing the top 10 Bris speed ratings.

It's like what All-Ways software does with listing the top 10 Bris speed ratings. — Mark

Exactly!

I guess the question is after you have correctly predicted the odds on the top 3 or 4 horses in a race, and knowing the top two odds horses win the lions share of races, what do you do with this? Is this your contender list based on your predicted odds? Or do you have a separate contender list from which you play your lowest odds contender(s) or conversely do you work from the other ends of your contenders list using the predicted higher odds contender(s)? I guess I'm presuming there are three or four contenders in a race.

What I do is first look for overlays in the top 3 or 4. If I find a decent one, say my 7-2 line 3rd horse is 8-1, I'm really interested. If noting jumps out, I look at the lowest line horse and disparage it. Why can't it win, what's negative in its record, why can't it win? Often, the lowest line horse is just "terrible" on paper, all of its good numbers coming from a turf marathon on yielding grass a year ago in a stakes race and today it's in a 5.5F claimer on dirt. I may mentally add a half point for each major negative I find, say taking the horse from 9-5 to 3-1. Or just say "no thanks" and move on.
Since only the last line is used, I also double check to see if a potentially top rated horse got left out due to a bad last race that is really excusable.

If I have a 8-5 top horse, and the next horses are bunched, say 4-1, 9-2, 5-1....I might consider a part wheel over them, or, if the top horse is suspect, I might bet 2-3 others to win or W/P. The handicapping begins once the line is ready. Funny, it is much like the process Beyer outlined in his chapter "Putting it All Together" in his book Picking Winners," and just yesterday dnglfnk at PA talked about HE uses that very chapter in his process.

Tom, thanks for sharing.

Tom, thank you for that explanation

I guess the question is after you have correctly predicted the odds on the top 3 or 4 horses in a race, and knowing the top two odds horses win the lions share of races, what do you do with this? — Gregory Byrnes

Wait.

Hope I didn't come across as terse. Just getting back to this.

I think you have to give prime consideration to the top 3 UOL horses. That should get a lot of winners. Then find your most successful contender process. Unless you feel strongly about one of your other contenders, you know that one of the three UOL low odds horses that is on your contenders list will win a lot of races. Then you have to look at which of these contenders will be aided by the pace setup. My contenders list needs to include both early and late horses. Will the race run early, late or chaos (late)? That's what I am doing.

Tom: what is the current points classification system? In the instructions it is 10-6-4-2-1 for horses rated 1-5 in the following categories: last race speed, average class or trainer win % if you have premium, prime power, early pace and late pace. For Aqeduct you changed it to a 74 point total with more emphasis on prime power and less on average class with early pace and late pace left out. Was that system just for Aqeduct or is that now the system for all tracks? Thanks for your help and a great tool!

It was just for Aqu, based on ongoing records, but it has stood up nicely at other tracks. I am going to re-evaluate for Bel as I get more data. I am just starting to handicap Bel this week.

Tom: your ultimate odds line was all over Early Voting for the Preakness. Nice work!

Tom is a very intuitive player. I have no idea how he does at the windows but I have to believe he's a winner

Lawrence here. I'm a long time fan of Dave's products and a lapsed HSH user. I've been away from handicapping during Covid, but I've decided to have another go at it as racing gets started for the season at Hastings Park, my local bullring. Honestly, I was inspired by the elegant simplicity of Roger LeBlanc's Instant Value Handicapping as outlined in his book, The Punter's Tale and on his YouTube channel.

Alas, this isn't a promo for Roger's stuff.

I've been on a programming frenzy in my Excel-powered handicapping workbook and I thought I'd share what I'm working on in an effort to give something back...and get feedback on if I'm crazy or not. It's been a long couple of days typing away!

Anyway, I was inspired by Tom's UOL to try and see if I could create a "master consensus" - a collection of ratings from diverse handicapping systems and approaches - in a search for insight into a race by considering it from multiple perspectives. (This approach is inspired by Mark Cramer's "informed minority" concept.)

To that end, I've created a group of what - for the sake of simplicity - I'll call "odds lines" based on the output from the following systems:

- Tom's Ultimate Odds Line
- Dave's 1-2-3 System aMLQ
- Pandy's Formula Pro System
- Pandy's Real Time Handicapping System

I've programmed each of these into my handicapping Excel spreadsheet and created a tool to compare the various "odds lines" to look for consistency and aberrations. This way, if one system misses something, another might highlight it. You can see the screenshot I've attached for a view of what I'm talking about. I've just finished the UOL+CSR Master Consensus page it so it's totally untested.

That said, it was the race pictured that inspired my thinking about building a "consensus". The winner - Codetown in the 8th race at TAM on 29 April 2022 paid $18.80 to win - was the 5th choice on my UOL line (that I've adapted to include a score for the RDSS time-decayed "Composite Speed Rating" that's come up once or twice on the forum) and 7th choice on the aMLQ when I 1-2-3'ed the race as a check for what I'd missed. On a lark, I checked the horse in my sheets for Pandy's systems only to find that that Codetown was the 2nd choice in both...and 2nd only to a LeBlanc IVH "false favourite." Ugh. I was frustrated and shocked. The info was just sitting there waiting for me to look at.

I got to work. I collated all of the system odds lines into one at-a-glance sheet. I included some measures of central tendency calculations to highlight when there were meaningful disparities between the lines of the different systems. And that's where it stands now.

I'm going to try it out and see what happens. Will it point me to other horses that the UOL overlooks? Will it indicate when the UOL has a "bad line" because it can't account for more off-the-beaten-track factors such as workouts or blinkers off (which has been a thorn in my side recently).

Perhaps someone on the forum can think of how it might be useful. I'm not sure I have a line or 4! - that I can bet with but it's got to be good for something, no?!

I'm trying to figure out how to share this photo. I'll see what I can come up with because the "Image" and "Upload Files" ways aren't working at the moment. User error, most likely.

Done and dusted!

Just remembered that Dave's Organize Your Handicapping seminar has info on how to manage information from multiple source. I'm off to have a watch of that right now. Maybe it will provide some insight on how best to use my Master Consensus sheet.

I'm trying to figure out how to share this photo. I'll see what I can come up with because the "Image" and "Upload Files" ways aren't working at the moment. User error, most likely. — Lawrence

Before I tell you how please I am to see you here - just pretend I didn't say that - let me tell you the easiest way to upload a file. OH! You figured it out!

I LOVE what you are doing.

Let me know how I can help.

BTW, I will predict that what you find from your spreadsheet work will work but not the way you think.

I'll guess that you will find that the obvious horses are what have been dubbed BOLOs by the locals. Can't recall at the moment whether it was @RanchWest, @Biniak, or @Steven, but it was definitely one of them

A BOLO is a BET AGAINST LOW ODDS horse.

Congratulations on some nice work.

#BOLO

Can't recall at the moment whether it was RanchWest, @Biniak, or @Steven, but it was definitely one of them

A BOLO is a BET AGAINST LOW ODDS horse. — Dave Schwartz

I think it was Biniak who dubbed it BA LO.

LeBlanc IVH "false favourite." — Lawrence

Have you found this to be effective? On Sunday, 5/22/22, I used a BRIS par drop of greater than 3 and and a morning line of less than 3.01. That did not pan out very well on the day. Of 19 horses, there were 7 winners. There was only one dropper of greater than 12 par points and at a drop of 21 that one lost on 23 days rest. 6 of the 7 winners were dropping either 5 or 6 points. There were only 3 losers dropping 5 or 6 points.

The payoffs ranged from $2.50 to $6.80, so maybe that's just something to live with since 12 of 19 lost.

Do you have any suggestions as to a way to get this to work better? There was a winner that dropped 12 points. I looked at days away and I didn't locate anything effective. Should I only look at the massive droppers? Or was my sample size just too small?

Or do you think this is only useful as part of an odds line?

Correction, 17 horses qualified, not 19.

3 at a 4 point drop all lost.
6 of 9 at a 5 or 6 point drop won.
3 of 4 at a drop above 6 points lost.

I may have found an important stat. 5 of 9 horses that were #1 in prime power won. So, ones not #1 in prime power had 2 of 8 as winners. Unfortunately, 2 of the 4 droppers above 6 points were #1 in prime power, but only 1 of those 2 won.

Yes, the false favourite criteria that LeBlanc outlines in his book have proven to be very effective. I've subscribed to his reports and - of the 3 races that he highlights on both Friday and Saturday - the favourites lose about 80% of the time.

I've only been playing along for the past 3 weeks or so but Roger has been posting updates on his lazybettorusa blog and on his YouTube channel.

Now that I have the false fave criteria programmed into my workbook, I can load up a race, see if the favourite looks false and pass or dive deeper. This helps me focus on the races that provide me with the best betting opportunities.

In terms of a BRIS par-based approach, I've never tried it. I do know that Dave has promoted a "strength of field" idea in lieu of pars, seeing as the strength is based on the actual horses in the race and not the ones that raced in the past. The strength of field calculation criteria are in P&P - both 2012 and 2022.

Thanks, I have Dave's strength calculation from one of his products. I will take a look. My apologies for dragging out my research one post at a time.

Update on my Master Consensus project using Tom's Ultimate Odds Line and various other systems.

I was finding that I had so much information to sort through - I had info from essentially 7 systems and methods - that I wasn't getting a clear signal. So, inspired by Dave's Figure Out What Wins seminar, I thought I would create a model for my local track - Hastings in Vancouver - to see if anything jumped out at me from all the information I had in my spreadsheet.

That was a stroke of genius! I found 11 factors - and some non-obvious ones such as some factors based on Bob Pandolfo's systems (Stamina, Form), a time decayed race class level one I homebrewed, Factor W Av2, and a "Late Factor" I picked up from somewhere - each of which hits at least 30% winners with rank of 1 and gets over 70% of winners in the top 3. Not to shabby.

I created an object using those 11 factors, used my aML object to figure out where to draw the line between the low odds Reynolds 3 contenders - they need a 111, 112, or 122 - and my 5-1 to 20-1 "sweet spot" contenders that can qualify with a Reynolds 2 - 11, 12, or 13, and tested it out against 7 races last weekend.

Still early days, but the method is proving a great way to find contenders. In my 75 race model - that's all the races they've run at HST to this point in the season, I average 4 contenders a race and get 90% of the winners. I'm using the 123 System worksheet to determine the value of each race - I will play if the Race Value is over 30% - and to optimize my betting.

I'm looking forward to playing this weekend to see if the methodology holds up.

Sounds like a very interesting project! I like that idea of a time decayed race class level.

Thank you for walking us through your process Lawrence. Very clever use of combining some software to get strong results. Please keep us updated.