In the last two issues of PPM, my old schoolmate and colleague Roman Kozlovski has drawn from his personal experiences of more than 40 years on the punt, forwarding to readers his five most valuable lessons learnt during this period.

In principle I fully concur with most points, though I am certainly at odds with his stance on the issue of value – to the point where I think he’s got this one horribly wrong.


The whole issue of value is inexorably tied in with market dynamics and something known as true odds. I have fleetingly touched upon these points in several of my previous articles for PPM but have decided to collate a lot of information from those articles and put this together as a definitive piece to help punters get a grasp on the whole issue of value in the racing domain.

Value in racing is contingent on something known as true odds, so we need to initially examine and understand what exactly the term ‘true odds’ means. There are always two types of odds – odds that are on offer and true odds. It is critically important to long term success that the player is fully aware of the margin between the two. True odds can best be described as when the payout to the bettor exactly matches the probability of the event happening. Let’s look at an example of true odds in the casino environment.

A standard roulette wheel with a single zero has 37 numbers. A player who stakes a bet on one single number has one chance of winning and 36 chances of losing; therefore, the correct payout on this event should be 36/1 ( true odds) but the casino only pays out at 35/1 ( odds on offer),  thus keeping one single number (the zero), in its favour at all times.

This equates to a percentage advantage for the house of 2.7 per cent on this bet type. If the player places a bet on either red or black, high or low, odds or evens and think they are getting a fair bet, then think again. They have 18 chances in 37 of collecting, which equates to a percentage of 48.65 per cent – therefore the correct payout should be 51.35  per cent but the casino only pays out at 50 per cent, thus holding a 1.35 per cent advantage over the player.

These shortfalls to the player will kill off all long-term gamblers in any casino environment. Although the edge is only slight, it will have a crushing event in the fullness of time fleecing gamblers of everything, but most cannot comprehend the mathematical futility of their endeavours.

True odds and probability are easily calculated in the casino environment simply because the odds on offer are static and the winning and losing chances in each game are clearly defined. In racing, the winning and losing chances can be a somewhat grey area and the odds are constantly changing. So when it comes to the racing arena, true odds evaluation is in essence a purely subjective evaluation.

In Roman’s July article he asks, “How can you prove the horse you priced is value at 2/1 or 3/1 or 5/1? The fact is you cannot pre-race and it is only after 100 bets on a 100 horses you priced at 2/1 that you can tell if you’re right.”

Well, in answer to this you don’t have to! If I price each runner in a race this is merely a rating of each runner’s chances expressed as odds.  Value doesn’t enter the equation unless there is a real time market running for comparison. If I rate a horse at 2/1 and I can freely get 3/1 or better, then my horse becomes value on the strength that if I have rated the race accurately, then I have given that horse a greater winning chance than the market is reflecting.

The proof would come per way of performance history. If I  had demonstrated over a long period of time my race assessments were accurate in terms of profitability, then given the nature of the arena we are operative in, I would deem that history shows that I can consistently isolate value by the comparison of my rated markets against available odds. So on the basis of that, pre race, I can say my backing a horse at 5/1 is value against my own market assessment of 3/1.

The very core of successful punting is the ability to accurately assess the genuine winning chances of any race then frame an accurate market around that assessment. Personal markets are compared to real time odds on offer and then the overlays are backed.

If you rate a horse a 4.00 chance and best odds on offer is only, say, 2.80 then this is considered an underlay and wouldn’t be backed because it’s perceived as poor value. This self assessment and comparison format is the modus operandi of every successful punting syndicate around. Value is in the overs and we MUST obtain value; it is the very heart of professional punting operations – isolating runners whose price exceeds their calculated winning probabilities. Exploiting the discrepancy between accurate pricing assessment and odds on offer is the only road to long term profitability.

The wealth of the world’s casinos is based on something known as Bernoulli’s Law, which is very powerful and basically implies that although gamblers may experience short term success, casinos can ultimately be assured of winning in the long term, or in other words, gamblers cannot win long-term whilst gambling against a bias.

Every casino game is biased against the player; the bias is the house edge, which is created by not paying winners out at true odds. If we transfer this principle to the horse racing arena, then, if our pricing assessments are correct and we accept payouts at less than our calculated pricing by backing underlays, then the casino principle will assuredly grind us into the ground long term. Roman’s article heading of ‘The Perils of Chasing Value’ needs to be amended to ‘The Perils of Not Chasing Value’.

Value is the single most important thing, no matter what you bet on. A horse can have a compelling formline and be an undeniable winning chance but whether it constitutes a good bet or not is solely contingent on the odds on offer. A regular stream of winners per se is not enough to guarantee long term profitability in this industry.

It is not the number of winners but return on investment that counts. Roman states in the article, “In the finish I simply decided I would accept the market’s assessment”  but this mindset of a winner’s a winner will not hold up long-term if you continually go down the road of accepting any odds on offer about your selection. It flies in the face of convention of nearly every successful punting model known.

If you’re seeking an endorsement about these points I’ve raised then just Google in Alan Woods/Bill Benter who formed the most successful betting syndicate in the world some years back. At the height of their powers they took more than $40US million per year out of Hong Kong racing. Such was the impact of their betting operations that the Royal Hong Kong Jockey Club had to place restrictions on certain aspects of their wagering because it was considered they had an unfair advantage.

Their whole operation was built around value and betting the overs. I wonder how they would have reacted if someone had tried to explain ‘the perils of chasing value’ in the midst of their onslaught.

If we look at this from another angle, skilled analysts and professionals are merely exploiting market anomalies. Any racing market is just a reflection of public opinion, but it is the wagering public themselves who create these anomalies by both over betting and under betting certain animals in the race.

Roman is certainly right in his statement that the marketplace is pretty much spot on in its pricing assessments. These markets are framed by full-time professional form analysts, but we need to be mindful that each market framed will have a built-in bias. This bias is the bookies or TAB’s advantage, so technically we are somewhat at a percentage disadvantage even before we bet.

Once the race is open for public wagering however, the market becomes dynamic and public betting trends then create anomalies and windows of opportunity for the shrewd punter to cash in.

Let’s look at a hypothetical situation and assume there is no bias built into the market, where we have a race of four starters each carrying the same weight and there was nothing between them on current or historical form, neither runner had any advantage over the other. Each runner would then be perceived as having one chance of winning and three chances of losing - so the true odds of each runner would be 3/1.

We also know that each runner is ridden by the top four riders in the state; however, we’ll also assume this race is in Sydney at the height of the Darren Beadman/John Hawkes dominated era in which each of their runners attracted huge public support. The weight of public money would then push this runner’s price down to less than its true odds of 3/1 and thus a market anomaly has been created by one runner being overbet to the point where its price has now become less than its actual probability of winning.

Shrewd operators will now deem this a no-go zone. As certain runners are overbet in the marketplace this will inflate the prices of other runners and create overlays in the race, which in turn creates the value situation we are talking about. 

On modern betting exchanges, (Betfair) the exchange is merely a meeting of minds and opinions on which prices are formed. There is no obligation to accept the offered price and users are free to offer or accept whatever price they wish. This should lead to what is known as market efficiency.

When we speak of efficiency in terms of markets we refer to information markets, with prices reflecting all available information. The detection of any market inefficiency denotes that a security or bet has either been overvalued or undervalued. Again the ability to detect these inefficiencies or anomalies is the real key to long term success in horseracing.

I cannot recommend strongly enough that any serious punter should learn the art of market framing in horseracing. Again just surf the net; there are a multitude of sites available that can set you in the right direction. It can be as simple or as complex as you want to make it.

The art of horseracing betting is possessing the ability to identify horses that are both over and under the odds, not simply picking winners. It is so much more about odds than it is winners. If a bookmaker can get you to accept $3.00 about a horse that should be $3.50 then he will assuredly beat you in the long term. Good things come and go, but percentages go on forever.

Obviously horseracing is not an exact science, but skilled analysts can do the form to the extent of getting a very close approximation of every runner’s true chances. Once and only when we get to that point can we start talking about true odds in racing.

So I close this out by saying take any odds about a wager at your own peril. Those who do this on a consistent basis are no better off than those ill-informed gamblers who flush their money down the toilets at casinos. It has never worked in that environment and it sure as hell won’t in horseracing.

By Ken Blake

PRACTICAL PUNTING – AUGUST 2009