Professional Level ROI

Professional Level ROI

Recently, a long-time subscriber emailed – asking (in brief):  “Hey Gary, what is the target ROI that pro handicappers believe to be acceptable when working consistently?

Well now . . . that pretty much is the $64,000 question.  The answer to, and an extended discussion of that question follows. 

The answer, of course, is relative to your Bankroll size and the average Pool size into which you place your wagers.

As well, you need to also answer the simple, yet slippery question, “What do I require from this game?”

I could offer a number that would be more-or-less accepted by most serious players:  +10% ROI  (and let’s restrict the discussion to win-betting only).


But – wait . . . now we’ll get into it a bit more in depth . . .

First of all – players need know what their projected ROI might be, and that can only be estimated from a history of actual wagers.

Look at your wagers in ‘groups’ – the last 50 wagers, the last 100 wagers, the last 300 wagers, and the last 1000 wagers. If you are using a consistent approach that doesn’t jump about as pertains to the wager decision-making – then each succeeding group increases your “certainty” that you will be able to project those ROI figures into the future.


Now . . .

If a player has a large Bankroll and is pushing, say, half-Kelly into the pools at a small venue – he will just be shooting himself in the foot.

His bet size relative to the pool size will cause every wager to pay less than it should/would have – had he not been involved in that pool. A vicious circle ensues, and his ROI spirals downwards.

So – let’s set preliminary parameters at:

  • Player has a minimum available bankroll that is 1/25th the size of the average total pool into which he wagers.     
  • The tracks he plays offer an average of 9 races a day each (from which he usually bets around half of them).
  • He follows 3 tracks per day.

Further assume that player is not a long-shot chaser, gets an average mix of pays at 2-1 to 15-1, and is carrying a project-able edge on the game of +5% – but his bets are restricted to around 2.5% of the pool size (because more will impact pools and lower his return for each wager).

Okay – with that we’ll compare potential results at 3 tracks . . .

Indiana Downs: This track might have an average win pool of around 30K in a non-stakes weekday race, so –

  • Player needs a minimum Bankroll of $1200 (1/25th of the pool)
  • His max allowable bet would be $750 (2 ½% of the pool).
  • If ½ Kelly reflects his 5% advantage – then 2 ½% of his bankroll is his actual max bet.  He would now need a $30k bankroll in order to make the max $750 bet in each race.

Say he finds 3 similar tracks to follow each day – and hits it full-time (which we’ll call at four 10 to 12 hr. days per week) . . .

This player would make 4.5 x 3 or around 13 wagers per day of $750 each. So he runs $9750 through the windows each day x 4 days = 39k per week = 156K per month.

If his ROI mirrors his edge of 5%, he gains $7800 per month.  A ten month year equates to $78K for this professional player.

Okay – but obviously the average ‘would-be’ pro doesn’t operate with a $30K bankroll, nor make $750 per race wagers. 

Scale it down by a factor of 5 and you have a $6k bank making $150 wagers and thereby making $16K per year.  

But – if using our originally suggested 10% ROI (which would also require a 10% advantage on the game) –  The real pro would then make around 150K per yr. and the ‘would-be’ pro around 32k per year.


Belmont: Let’s call the average week-day win pool at 150K, so –

  • Player with 1/25th of the pool has a min. bank of $6k. 
  • If he makes the same 2½% of pool wagers as the Indiana Downs player and can bet the max – that max bet would be around $3500 per race
  • Which would require, though, a bank of 150k.


Again – if this player can handle and find 3 such tracks, and make the same number of wagers as the Indiana Downs player – The pro will wager around 182k per month and make for his 10-month year right at 100k.  Mr. Would-Be (scaled down by a factor of 5) will make around 20K.

And if using the base 10% ROI figure?  This would give a real pro with deep pockets a shot at 200K per year, and the would-be pro at around 40K.


I have not mentioned rebates. There are the syndicates and “whales” out there who pretty much play only for the rebates and are happy with a break-even ROI, or small actual losses on their handicapping. Because they push enough money around – the rebates alone still offer them a shot at professional-level income.


Hong Kong: If you are of the William Benter, Alan Woods ilk – and play into pools that run in the millions – then even a 2% ROI could be extremely lucrative. 

With an average pool size of say 3 million – then one could wager 75K per race without unduly affecting the pools.  There are not many tracks like that worldwide, so far fewer races will be available. 

If there are only around 90 days/nights of racing in a year – 7 races per. That’s 630 races – so the player might send 47 million through the windows in a year.  At 2% ROI ?! – that’s about a million in profit per year.

So . . .

Barry Meadows and many of the ‘skeptical, probabilities boys’ will tell you that a consistent 10% ROI is great – and pretty much the ceiling for a professional bettor. 

My mentors, Doc Sartin, Jimmy Bradshaw, and Dick Mitchell – used to scoff at a figure that low.  But every player knows his own bottom-line truth – and that is often a bit lower than he might let on.

Counting the bad runs, the unforeseeable blue-moon losing streaks, and the recouping of major Bankroll hits . . . Counting the costs of doing business – the loss of income during sickness, vacations, burnout breaks, etc. . . .

Anyone looking to make their living from this game should not have long-term expectations exceeding (and this is my personal opinion – based on my own results over 40 years) . . .  +18% bottom-line net.


Then – my long-winded reply to the initial question at the top of this post can be boiled down to:

A professional player needs to have a bankroll large enough that a 5% to an 18% ROI will supply the level of income he requires – given the restrictions of the pool sizes, and the number of plays he can make per month at the particular tracks at which he wagers.


As  always – comments are welcomed.
Best of Fortune to you.


  1. Hi Gary,

    I have Dick Mitchell’s “Winning Thoroughbred Strategies”
    He recommends no ROI less than 20%.

    Thanks for all you do.
    Joe B.

    • Joe – I knew Dick Mitchell personally – good guy – and his information is valuable.When you say “recommends” a 20% ROI ? If it’s simply a recommendation, then why not 30% or 40?! I’ve been in the game a very long time – and gaining a long-term (meaning years running) ROI of 20% is difficult to achieve. The ones who have done it are as rare as hen’s teeth. Which isn’t to say it can’t be done –

        it can

      – but . . . it is no easy accomplishment.

      Warm regards – Gary

  2. Dave Henry

    Had to look up the Kelly criterion and saw this from The Hackensack: “…With the half Kelly bet, your probability of temporary loss is a quadratic function of the amount of loss. For example, you stand a 81% chance of losing 10%, a 64% chance of losing 20%, a 25% chance of losing 50%, etc.” But other than trying to decipher some advanced math formulas, I could really appreciate the logic in this article. Have always preferred venues with the larger handles – just makes sense. And now you’ve made it clear why that is, as well as how to determine maximum bet size. Your background in commodity trading is paying off. Also, betting 2% of your bankroll into any given pool seems safe enough. You’d need a losing streak of 50 bets to go broke. And if I ever did that I think it would be time to consider taking up a different hobby.

    • Dave – thanks for the comment. How hard a player can press Kelly is also a function of his hit %. The higher your hit rate (given that your bottom-line is positive) the more (% of Bankroll) the Kelly Criterion will all you to wager. Even with a higher ROI – a profitable low hit rate method (i.e. one with a higher average mutuel) will still need to wager less (re Kelly) than the higher win % bettor. – Gary

      p.s. I have sent you a pm with a tool you will find of interest.

  3. Ted Barrow

    Would like to take advantage of 40 % off if still available.

Leave a Reply

Your email address will not be published. Required fields are marked *