Legendary punter Don Scott: Do his winning methods apply today?

Most serious Australian punters have heard of the legendary form analyst and author Don Scott.

Don Scott was an innovative and very successful form student and punter (at least for a while!) who published his first book Winning in 1978. Scott is recognised as the pioneer of weight ratings, and many current ratings programs refer to his work. He then published other iterations of Winning. I have a copy of The Winning Way published in 1982. Don Scott’s original work is 44 years old this year, and the punting landscape has changed during that time due to the influx of data, the wide use of sectional times and the computer age.

The basis of Scott’s weight rating scale is his 15 Maxims. They’re listed at the end of this article.

Don Scott’s time versus the current day

Many of Don Scott’s maxims are embedded in the process of “doing the form”  in 2022. The difference between 1978 and now is that most current ratings programs have an inbuilt process, using an algorithm, to determine a rating for the horse in each race. This algorithm will differ from one program to another.

A bigger difference between now and 1978 is the wider use of sectional times (and how they are calculated) to determine the “rating” for the race and the horses in that race.

Which races?

Don Scott also talks about the right and wrong races to bet into. Some of his “wrong” races are where the going is on the worse side of slow, early two-year-old races, and low-class races.

Some of these rules I do not follow. I bet into maidens and BM58s if I can identify horses with factors that are worth a bet, and I bet on slow and heavy tracks – although I do steer clear if the weather on the day is dubious.

Scott also believes in the “odds-on, look on” theory, which I do not.

Betting to value

Don Scott was an early adopter of pricing your own markets and then trying to find overlays. This theory is very much the basis of most punters today. He uses fractional odds in his books, as they were in use at the time, to demonstrate his overlays and what he would need to bet.

The approach, basically, requires the punter to know what a horse’s true odds are (over the long term) and betting to those odds. For races already run, this is essentially the starting price (SP), as the SP has been shown to be very accurate.

Using my database, I looked at various SP profiles and what the true odds are (remembering that the SP is not a 100% market, so will not reflect a totally accurate probability). Here’s two examples:

• $2.00 SP (50% chance): These win at 45.5% strike-rate, or $2.20 (sample of 2440 horses).
• $2.50 SP (40% chance): These win at 36.4% strike-rate, or $2.74 (sample of 6932 horses).

Betfair SP runs very close to the “true” odds over the long-term, as it is very close to a 100% market:

• $2.00 BSP (50% chance): These win at 50.2% strike-rate (sample of 1543 horses)
• $2.50 BSP (40% chance): These win at 39.9% strike-rate (sample of 1940 horses)

The accuracy of the Betfair SP market is clear.

This to me is one important side of the betting “market”. It is why I talk about whether I am betting SP or Betfair SP on my bets over the long-term.

Your betting profile

In my opinion, the other side understanding your profile as a bettor.

What is your long-term strike-rate? This not only impacts the probability of losing and winning runs, but also the odds you are taking for your bets.

I also need to know and understand how my ratings and markets are performing. What happens if the horses I am marking $2.50 (40% winning chance) are only winning 30% ($3.33) of the time? With a big enough sample size, this can create a losing scenario. I want to be confident that my pricing is close to its true winning chance over time.

I also understand which prices in my market I have an ‘edge’ in. I know that I have a greater edge in horses in the market – rather than those longer than $10 SP. I know this as I collect data on it and thoroughly analyse it in intervals throughout the year.

As Betfair SP is so efficient, it is also why betting movements can influence my markets. I can blend the market with my prices. I believe the market is efficient and long-term is the best indicator of “true odds”. For certain profiles of horses, if they are firming on Betfair and then therefore the bookies, I am happy to adjust my market to match these. It is why prices that incorporate Top Fluc are great, as if they keep firming you are beating the SP. Over the long term, that is positive.

Using data to test theories

We spoke about the differences between 1978 and 2022. One big difference is the amount of easily-available data.

I can run thousands of queries in my Ratings2Win database to see if my hypothesis on “edges” are positive or have no “edge” in the market. Those who have asked me questions about data know how quickly I can get it.

This allows me to check the validity of long-held theories. For example: you may have heard of the “mares in-form” gem that gets thrown around. But what does the data say?

Horses who have won their last start:

• Geldings 14.4% strike-rate, -12.43% PoT (39,779 horses) 
• Colts 13.7% strike-rate, -15.59% PoT (4234 horses)
• Mares 12.4% strike-rate, -17.77% PoT (14,188 horses) 
• Fillies 12.4% strike-rate, -15.58% PoT (17,010 horses)

Horses who have won their last two starts:

• Geldings 13.7% strike-rate, -12.43% PoT (16,022 horses) 
• Colts 12.6% strike-rate, -18.0% PoT (3376 horses)
• Mares 11.9% strike-rate, -17.82% PoT (3855 horses)
• Fillies 11.1% strike-rate, -16.32% PoT (12,235 horses)

There is nothing in this data to say that mares or fillies in form are a betting advantage. The data says that geldings who have won either their last start or both last two starts are the best betting horses. Even when you show this data to some who believe in this mantra, they will start with “Yeah, but…”

I throw around many different hypotheses in my head each day to see where there are “edges” in the market. The issue  is the amount of data you need to make it a “valid” sample size, versus what the current market is doing. If you are using data back to 2018 to get a large enough sample size… is the market today still the same?

So what of Don Scott?

As you can see, many of Don Scott’s idea are still around today. Some I check to see if current data suggest it is still an edge, or if his theory is right. The database I use factors sectional times into the ratings, as well as other things that Don Scott didn’t. But in identifying a potential rating for what that horse runs today, I do factor in some of Scott’s maxims, as well as other things today’s data and markets are showing.

Scott is also just one of many authors who have published work on the topic. I have also read and consumed many different horse racing, financial, mathematical, psychology (decision-making theory) and sports betting books, podcasts and articles to shape how I bet in today’s market. It all helps shape my opinion and betting profile… but fundamentally, my edge in the market is data analysis.

My process in decision making on betting is quite rigid. It stays the same day-in, day-out, but my ideas are fluid depending on whether they can be validated by data to show an edge. The two questions I am regularly asking myself are:

1. How are my ratings / prices comparing to their actual winning probabilities; and
2. How am I comparing to the market?

There are many ways to win in today’s market. You just need to know what your niche and edge are.

Don Scott’s 15 Maxims

1. All factors that affect the rating of a horse are expressed in kilograms.
2. Extra weight is a disadvantage to a horse. 1.5kg of extra weight disadvantages a horse by one length.
3. Some outside barrier positions are a disadvantage to a horse. This disadvantage can be expressed in kilograms.
4. A less skillful jockey is a disadvantage to a horse. This disadvantage can be expressed in kilograms.
5. Some horses suffer running disadvantages during a race. Most running disadvantages can be expressed in kilograms.
6. The form of horses rarely remains steady. Horses are usually either improving or deteriorating. This improvement or deterioration can be expressed in kilograms.
7. When horses compete on equal terms, their performance can be rated in kilograms, depending on their finishing position.
8. A horse has to be compensated in its past ratings for any relative weight disadvantage. This disadvantage is always a plus figure in kilograms.
9. The higher the kilogram rating of a horse, the better it is.
10. Any weight disadvantage retards a horse.
11. Past ratings can be produced by comparing horses through direct and indirect weight lines. Past ratings are necessary to produce future ratings.
12. Past ratings can be produced from my class and weight ratings tables. Past ratings are necessary to produce future ratings:
13. Past ratings or true performance ratings are produced from losing margins, jockeys’ relative skill disadvantages, relative weight disadvantages and the class and weight rating tables.
14. Future ratings are produced from past ratings. Future ratings contain an allowance for barrier position disadvantages, jockeys’ relative skill disadvantages and relative weigh disadvantaged.
15. All factors that affect the rating of a horse are expressed in kilograms.