Importance of Previous Starting Prices

There are many different aspects of a horse’s form that help to determine its winning chance in an upcoming race. Factors such a previous finish positions, margins, class of race, distance changes, runs from a spell, days between runs, win and place strike rates and even at a more sophisticated level, previous performance ratings are a key part of the form analysis process used by many punters.

However, one aspect that receives little attention in the mainstream commentary we read and listen to about races are the horse’s previous starting prices.


1. What are Previous Starting Prices?

If you look in a horse’s form you will find all the details you are familiar with about each past run, but then somewhere in the text you will also see a market price i.e. $5.00.What this tells is the horse started $5.00 in the betting market for the race, based on the official price methodology employed in that State (which is typically a sample of Corporate Bookmaker prices.)

You may also see other prices shown, which represent the horses price fluctuations in official betting, which is generally the 30 minute leading up to a race. If the form data shows $4.00, $6.50, $5.00… this would indicate that the horse opened $4.00, drifted to $6.50 before being backed and firming into a starting price of $5.00.

For example, below is a copy of the Racing Australia web site form guide for the Darren Weir trained Black Heart Bart. His betting market prices for each run are highlighted in yellow.


2. What do Previous Starting Prices Indicate?

The previous starting price of a horse is a measure of its expected performance in the race. A horse that starts $2.50 in a race is expected to perform much better than a horse that starts at $10.

Historical data shows that $2.50 chances win approximately four times as often as $10 chances and have a much lower median beaten margin.


3. The Concept of Expected vs Actual Performance

When most punters study the form, they are almost solely focused on the previous actual performances of a horse.  That makes sense doesn’t it?  If we want to predict how a horse might run today, then how it has run in the past must hold the answers.

However just as important to assessing the chance of a horse in an upcoming race is the expected performance from its past runs, especially recent runs. There are any number of factors that can affect a horse’s actual performance on the day and while that is important, we shouldn’t lose sight of what each horse was expected to do.

The easiest way to demonstrate this concept is to think of your favourite football code. If a top of the table team meets one near the bottom, then all other things being equal, your expectations of performance would see the higher ranked team winning and perhaps easily. With the explosion of sports betting we can now see the precise expectations in the shape of betting market prices, as we can in a horse race.

Despite the pre-game expectations, let’s assume the top ranked team has an off game and the bottom ranked team performs above expectations and actually wins. You know that if the teams were to meet 100 times then the top ranked team might win 90 of those games, but on this particular day, that 1 in 10 chance of the bottom team got up.

What if those two teams were to meet the following week, which would you favour to win? If focusing only on previous actual performance you would have to conclude that the bottom ranked team is favoured to win again.However, we also have a natural instinct for the overall ability of football teams and are able to put upset results (i.e. that don’t match the expected performance.) into perspective.

What this means is that while you may consider the chances in the re-match to be closer this time, you would almost certainly still favour the top ranked team (clearly) to turn the tables on their last match and win.

This same concept applies to horse racing, although it involves a more complex structure and subtle judgements compared to matches with just two competitors that you are intimately familiar with.

The key point is that:

“The chance a horse has of winning an upcoming race is influenced by both its actual past performances, but also its expected performance in those runs.”

 Let’s take a look at some data which demonstrates this point.


4. Analysing Horses by Their Most Recent Starting Price

The table below shows 5 years’ worth of metropolitan racing with each horse analysed by nothing more than their last run starting price, excluding runners first up from a spell (where the last run was 56+ days prior.)

Last Start SP$ Runs Wins SR% POT%
$1.10 to $1.90 3416 767 0.225 -0.043
$2.00 to $2.90 9588 1733 0.181 -0.04
$3.00 to $3.90 12055 1784 0.148 -0.074
$4.00 to $4.90 14115 1924 0.136 -0.059
$5.00 to $5.90 10548 1278 0.121 -0.091
$6.00 to $6.90 9686 1086 0.112 -0.085
$7.00 to $7.90 8776 909 0.104 -0.107
$8.00 to $8.90 7951 798 0.1 -0.086
$9.00 to $9.90 6809 607 0.089 -0.145
$10 to $12 14686 1270 0.086 -0.104
$13 to $16 14104 1104 0.078 -0.092
$17 to $21 14812 983 0.066 -0.12
$26 to $31 10434 562 0.054 -0.135
$34 to $51 5822 275 0.047 -0.12
> $51 5037 194 0.039 -0.145

Notice the way that the strike rate for chances in an upcoming race flows in direct relationship to the horse’s previous starting price.

Remember that these figures don’t consider where each horse finished last start, what the class of race was, the distance or any other factor is related to today’s race.It’s purely the relationship between previous starting price (or expected performance) and strike rate in today’s race.

It allows us to draw a very powerful conclusion:

“On average, the shorter a horse started at its last run, the greater its chance of winning an upcoming race.”


5. Consider the Class of Last Start

While the above table of data demonstrates an important point, you can’t put previous starting prices into proper context unless you also consider the class of those runs.For example, a horse that starts $1.90 in a Provincial Maiden is entirely different to one that starts $1.90 is a Listed race.

In the table above, you saw that horses with a starting price of less than $3 at their last start had the best strike rate and betting return.

Let’s take those horses and compare their success in the upcoming metropolitan race based on the region of their last run.

Last Start SP$ Runs Wins SR% POT%
LS Metro up to $2.90 5603 1420 0.253 -0.022
LS Prov  up to $2.90 5714 907 0.159 -0.057
LS Country up to $2.90 1687 173 0.103 -0.109

This table shows how important it is to understand the class of race with previous starting price.The total group of horses that started less than $3.00 at their most recent start have a 19.2% strike rate, but those coming off a metropolitan run at that price have a 25.3% SR, while those coming off a Country race have just a 10.3% SR.

Remember, these stats totally ignore how the horse actually ran at its last start!


6. Analysing Last Start Actual vs Last Start Expected Performance

Let’s create three main categories of horses that we can compare, based on their actual and expected last start performance.

Last Start Actual Performance:

  • Ran very well = Won or finished within 1 length of the winner
  • Ran fairly = Finished 2.5 to 3.9 lengths from the winner
  • Ran poorly = Finished 5+ lengths from the winner.

Last Start Expected Performance:

  • Expected to run very well = was less than $3 SP at last start
  • Expected to run fairly = was $7 to $14 SP at last start
  • Expect to poorly = was $26 or longer at last start

You will notice these groups don’t cover every horse in the population. I’ve simply taken horses at the top end, bottom end and representative middle group so we can make some clear comparisons.


Horses that ran well last start

(i.e. won or finished within 1 length of the winner)

Actually Were Expected Runs Wins SR% POT%
ran very well to run very well 9330 1933 0.207 -0.033
ran very well to run fairly 12964 1528 0.118 -0.091
ran very well to run poorly 1905 141 0.074 -0.125

Notice how the strike rate of these horses varies depending on how they were expected to run last start.

They’re all coming off good form, winning or finishing within a length of the winner at their last start, but those that were expected to run well have a 20.7% at their next start and just -3.3% POT, while those that overachieved as they were expected to run poorly last start (i.e. they started $26 or longer), have just a 7.4% strike rate and a betting return almost 4 x worse at -12.5% when they go onto their next start.


Comparing last start over and underachievers at their next start

Consider the following tables which compares those that overachieved at their last start (i.e. they ran very well against an expectation of fair to poor) with those that underachieved (i.e. they ran fair to poor against an expectation of running very well.)

Overachievers:

Actually Was Expected Runs Wins SR% POT%
ran very well to run fairly 12964 1528 0.118 -0.091
ran very well to run poorly 1905 141 0.074 -0.125

Underachievers:

Actually Was Expected Runs Wins SR% POT%
ran fairly to run very well 1036 155 0.15 -0.092
ran poorly to run very well 544 76 0.14 0.09

What this comparison shows is that horses with a fair to poor last start run, but started short in the market (<$3.00) actually have a significantly better next start strike rate than those coming off a good run (win or within 1L) that started at a middle to long price in the market.

It shows us that on average last start expected performance is more predictive than actual performance.


Let’s compare a group of horses that have very similar last start form

They are:

  • Last start metropolitan winners
  • Racing again in the same preparation in another metropolitan race
  • The upcoming race is of similar prize money (-$10k to +$10k)

As a group these horses attract plenty of attention from punters. They’re coming off a recent metropolitan win and line up again in a metropolitan race of similar standard.The prize money rule excludes those winners that are jumping sharply in class, for example from a midweek win to a Saturday class race or Saturday class race to a black type race.

Let’s break these last start winners down by the price they started in the race they won:

LS SP$ Runs Wins SR% POT%
$1.1 to $2.90 1100 380 0.345 -0.03
$3 to $4.5 1667 455 0.273 0.035
$4.6 to $6.5 1248 256 0.205 -0.046
$7 to $10 1276 221 0.173 -0.049
$11 to $17 882 119 0.135 -0.042
>$17 440 38 0.086 -0.182
TOTAL 6613 1469 0.222 -0.024

On last start actual performance only, this group of 6,600+ last start winners would appear very similar in the form guide.  However their chance of repeating the win depends very much on how they were expected to perform at their last start.

Those last start winners that were expected to run very well (less than $3 SP) have a 34.5% strike rate at repeating, while those with a last start SP of $7 to $10 have just half the chance of repeating at a 17.3% strike rate. Those that started longer have even less chance of winning again.


Assessing chances in an upcoming race

They key point is that when pricing horses for an upcoming race, don’t fall into the trap of just looking at actual performance. To price a horse’s chances more accurately you need to consider previous starting prices and how the strength of those races compares to the upcoming race. This is often the missing ingredient in the price assessments of a casual/hobby punter that tends to focus only on actual performances, compared to the more accurate markets of an experienced form analyst.

If you don’t consider previous starting prices as part of your assessment then you will end up pricing many horses too short on the basis of a good recent run and perceive them as big value, when history will later show that they weren’t.

You will also end up dismissing other horses too much on the basis of a moderate to poor recent run, when their previous starting price gives them a much better than average chance of regressing back to that expected level of performance. Whatever the case, it will lead to less accurate assessments and potentially less profitable betting decisions.


Wrapping Up

This data is just a snap shot that demonstrates the importance of previous starting prices in assessing chances for an upcoming race. There’s a number of different ways it can be broken down and different angles that can be analysed, each with their own insights. We’ve focused purely on the last start, which is by far the most relevant, but history prior to last start also plays a role.

The statistics represent overall averages for each group and there are of course an infinite number of variations in each scenario that might make one specific case different to the next and different to the overall average.

The data doesn’t mean that you should always ignore the form of horses that started longer in price and automatically forgive poor runs if the horse started short in the market. Each race and horse should be treated on a case by case basis.

They key point is to:

“Recognise how important previous starting prices are consider them as one input in your form study, especially if you are trying to frame markets for upcoming races.”

It could be the missing link that helps take your assessments to the next level.

Even if you don’t take that path, you can still use these insights to help better categorise the chance of horses in an upcoming race and find good betting prospects that appeal to you.


Related Pages

Introduction to Ratings

Ratings are numerical measures that reflect the quality of a horse’s past performances or other specific characteristics.

Read More

Betting with Ratings

In our last article we introduced the concept of ratings and how they are typically constructed. We also touched ...

Read More

Ratings and Prices

Ratings have become an increasingly important tool in the arsenal of a punter. This part of the series goes ...

Read More