Profit and ROI Question

low profit but a higher ROI.

If the edge persists over time (say 6-18 months), let the money filter down to more marginal profit opportunities.

Ok - first off: Are you making a profit already? Have you been making money on these bets and seeing your balance go up, or is this based on either history or a "filtered" selection of your results? If it's the latter then just try anything you like until you are actually making real money, then refine. Anyone can filter lists of "might have been" bets until a particular group of them shows an apparent profit. I'd wager that if you took a list of all 2y olds over the last 10 years(or whatever) and categorised them alphabetically you'd find a group of them showed a profit - that doesn't mean that horses whoes name begins with "G" are all value to back. Finding apparent patterns in existing data is easy because there are so many potential patterns that one of them is bound to come up, the trick is finding trends that project forwards. To do that you make a hypothesis, bet accordingly and then examine your results in relation to your hypothesis. "Filtering" these results retrospectively according to some criteria that you didn't account for in the hypothesis until you get the outcome you wanted will get you nowhere.

Assuming that you are already making money reliably then rather than cutting less profitable bets entirely it would be wiser to adjust your staking to reflect the expected return on your bets, whether using Kelly or a more basic means. In very simple terms - put more money on better bets. Not really rocket science is it!

ROI is hard to define if you end up betting against yourself. Do you define your ROI has profit/maximum exposure, or profit/exposure against each bet independently?

Stick the numbers for different levels of filtering into a Kelly calculator and see what the optimal Expected Growth is.

In general, unless having all your money in play becomes an issue, lower ROI is better.

A high strike rate with a low ROI is more sustainable long term than a high ROI and a low strike rate.

ROI say you start with a betting bank of 10,000 you turn over 100,000 and make a profit of 10,000 thats a 10%roi, you start with 10 turn over 100,000 and make a profit of 10,000 thats a 10%roi. thats why roi is a load of boll*ks imo takes no account of what you start with which is a very important factor

know all, I wouldn't quite say that.
When you combine ROI with the turnover and turnover ratio stats it becomes a lot more meaningful.

For instance, I know a guy with a ROI (margin) slightly better than mine, using a bank about 1.5x as large as mine, but he turns over more than 10x as fast, and so makes a massive profit.

I use the stats:

Average Balance    Bet Volume   
Daily Bet Volume   
Daily Turnover Ratio   
GP   
Net Trading Profit   
Profit on Turnover   
Market Turnover    (this is just the sum of the absolute values of individual NET market pnl)
Profit on Market Turnover

well i hope i have proven to some that ROI is very misleading unless you know a lot more, have the starting bank figures and the turnover figures

surely the most important thing is not ''starting bank figures'', ''turnover figures''
but how much money you are making, no ?

kenilworth
Date Joined: 04 Nov 05
Add contact | Send message
When: 02 Apr 11 00:36
Joined:
Date Joined: 04 Nov 05
| Topic/replies: 4,937 | Blogger: kenilworth's blog
surely the most important thing is not ''starting bank figures'', ''turnover figures''
but how much money you are making, no ?

They are all important.

But one is more important than the others.

Not really, if someone tells me they made £100k on betfair last year, and another guy made £90k, it doesn't say too much.

Maybe the first guy is using a £500k bank and the other a £5k bank.

Anything can be made to have more importance
if ridiculous examples are used.

Yes, but this isn't a ridiculous example. There are people making good money with a few grand and there are people who are using millions.

The % return on stakes the most important indicator i'd say.

That should be consistent if you are applying the same method.

I like to think that when it comes down to Fixed Odds Betting "Strike Rate" against Average Odds taken is somehow more Important than ROI,... or not ?!?!

I don't care much for ROI or turnover etc stats, because it becomes too time consuming and complicated when you're placing bets in and out of markets all day long, but the one that means the most to me is profit as a percentage of max drawdown, or plain old risk over the course of a month.

What is Max Drawdown?

How much you drop over a set time period, say a month, compared to how much you gain. So for example you could make say 2k over a month, and within that month have a max losing run of 2k, so your profit is equal to your risk, that's high risk as far I'm concerned. If you can manage to make say 4k with a max losing run of 2k, that's a good indicator of less risk or more profit as a percentage of max drawdown. The higher your percentage of profit, ideally 400% + (imo), the more you can think about raising the stakes.

Yes Trevh, I agree that's a very meaningful stat. For me ROMAD is more useful over a 1 year period.

For the purposes of deciding whether to scale up, it will be useful to look at the shorter timeframes you mentioned.

From 2010 I've become less and less cautious, which has lead to much more volatile returns. This is because I have a much larger cash buffer and because it is good for PC reduction.

the ROI was never meant for the likes of turning over large amounts of money on like horseracing or poker, its been adopted but its flawed and should never be used imo to decide if something is profitable unless all the facts and figures are known.

%  profit relative to bank roll is maybe a good way of looking at it?

Although, personally, I'm pretty sceptical of the concept of a "bankroll".

You could perhaps argue that, for pros, profit as a % of net wealth is a good indicator.

racingguru Joined: 06 Jan 01
Replies: 199 05 Apr 11 11:53   


Return on Investment (ROI) I see as % growth of your bank over a year and  Profit on Turnover or % return on stakes as the total profit/total stakes x 100 over a year.

The latter is the more important IMO as it defines your edge over the odds compilers. But the bottom line is total profit always has been and always will be.

Not according to the Investor.

I've always defined ROI as Profit / total invested. i.e. Total profit / total stakes. I thought that was pretty standard, although looking at definitions in the finance world there's an argument for racing-guru's definition.

These are all fairly meaningless metrics as far as I'm concerned. All you need to know is the edge on your bets in all applicable circumstances. Then it's just a question of optimising stakes and maximising opportunities. Who cares what percentage of your bank you're making or what your turnover is, if a bet has positive expected value then you should take it. Easy. Some of my strategies are better than others but that doesn't mean I cut the less good ones because they still make a profit. Assuming that a strategy more than covers it's costs (in my case that's things like server costs) you should use it. Why wouldn't you? It's all money and that's what we're here for.

So, robot, you enter any market if you believe there is ANY positive EV?

A real example of when I use ROI is deciding if I bet win only or EW.  By keeping records I can tell, if in the long run, I will get a better return by putting all my sake win only or splitting it by having a EW bet way.

Assuming it's sufficient to cover comm, costs and still return a profit - why not?

Nice post racingguru. Fully agree although the only thing all of us need to be careful with is over staking simply to ensure year on year return figs are simliar.

ty Trevh

Racingguru ,problem with the stats on Grade 1s i expect is there are so few of them, so even 3 years worth of results could be too small a sample to judge by (compared to Handicaps where 3 months worth of results could be a very reliable guide to how well you are doing).

Aye Robot - You said "Who cares what percentage of your bank you're making or what your turnover is, if a bet has positive expected value then you should take it."

If you are making less than 2% of your bank per annum then it is important because you would be better off withdrawing your money and investing it elsewhere. To that extent it is important what percentage of your bank you are making.

I personally agree with your other thoughts on meaningless metrics but that is just me, i'm sure other people may wel find them useful.

Ok - so it hadn't occurred to me that you might make even less money than you would in a savings account. Seriously though - 2% per-annum is totally dismal - almost worse than losing. If you're in that position give up!

I'd hardly describe that as dismal, seeing as you would be thrashing the vast majority of punters making 2% on cash.

Sure if that's what someone can expect to make long term, they should give up, but if you can find ways to profit (no matter how small), you can probably improve on that over time, as the fact that you're making anything at all puts you head and shoulders above the majority.

My annualised return for the Q1 2011 was 11%, but 2007-2011 average is about 100%

I made 2.70% in Q1 so annualised  growth factor is 1.027^4

I think we are getting confused with 2%~ ROI per bet and 2% increase of bank per annum.

If 'bank' means the same money being recycled again and again, my return on bank is sizeable--but if it means the total amount of value at risk w/ each bet considered independently, it's very small.

Incidentally there's another 'bank' I have periodically to dip into to hedge off multiples, jackpots and placepots. Enough of these win to make me a lifetime loser on here, over five years after being pretty much closed down by the online bookies.

I feel that if you consider drawdown w/out some measure of likelihood or historical measure of how often the money has been lost, you'd become too risk-averse. You might only arb.

Wicked Whisper, I'm talking about increase of bank.

I don't increase my bank, as I spend my profits. What else
are they for ?

Hello there Aye Robot. 2% was just a random percentage that I used. If you lived in Australia for example then you can use a higher percentage. On top of that you also need to take into account that it does require some work and effort to obtain your return from betfair, whereas if you stuck your money in an account you would not have to do anything at all to make a return. Plus you also have to factor in the security and the element of risk that you need to expose yourself to in order to obtain your return from BF. Taking all this into account an individual may feel that it's not worth doing it unless they can obtain a 10% plus return from their capital.

I suppose that for all the super successful people on here like yourself there are many more who are struggling and scratching around trying to make it work. How do these people determine whether it's worth continuing to carry on with what they are doing or to stop and think about doing something else instead. You said before "if you're in that position give up".

My point is that you need some barometer in order establish the worthiness of your betting quest and I feel that your percentage profit from your bank is a good barometer. Your are in a very lucky (although i'm sure lucky is not quite the right word as i would imagine that you had to work very hard) position where your return is miles above this threshold that it is not on your radar, but i believe that for many people the question of "would my money be better off elsewhere?" is much more relevant.

I do take your point U.A. - But I still think it's relatively easy to make that distinction without over complicated measures. I'd ask simpler questions like "am i earning more than I did in my job?" or "which direction is my bank balance going in". If it's in any way a marginal call after you've been at it for a while then you probably need to look elsewhere.

I think underlying some of this "analysis" is a desire to see how you're doing relative to others. That's pointless and silly. The only thing that matters (if you're not playing for fun) is how much money YOU make. What anyone else might be making and whether they're doing better than you is neither here nor there, there is no world championship for exchange gambling. You just fight it out at your own level in each market. Who's got the biggest ROI is just comparing willys.

kenilworth
Date Joined: 04 Nov 05
Add contact | Send message
When: 06 Apr 11 07:08
Joined:
Date Joined: 04 Nov 05
| Topic/replies: 4,951 | Blogger: kenilworth's blog
I don't increase my bank, as I spend my profits. What else
are they for ?


This will be the case for many people. I don't believe I am at capacity for my strategies, so I'm plowing much of my profits back to increase my trading capital.

This is how my 2010 GP was divided: Betfair 22% (comm, PC etc), Other beneficiaries 9% (software, hardware, interest etc.), me 19% (what I 'pay' myself), retained profit 50%.

aye robot
When: 06 Apr 11 13:07

[...]
I think underlying some of this "analysis" is a desire to see how you're doing relative to others. That's pointless and silly. The only thing that matters (if you're not playing for fun) is how much money YOU make.
[...]

I agree with that (with the caveat that a few other metrics can be important along with size of profit).
You should be making as much as you possibly can, regardless of how well others are doing. I don't think that means ignoring the performance of others though, as you can seek to emulate those making more than you.

It is often the case with companies that they react to hard times by finding ways to be more efficient. Obviously it makes sense to find ways to be more efficient before you are 'forced' to do so to avoid going bust. That's analogous to making as much as possible on here regardless of how well others are doing or how much you 'need' to make (while keeping a good balance in life).

Hello there Aye.

I think we pretty much agree overall with each other on this one. You say "But I still think it's relatively easy to make that distinction without over complicated measures" but i'm not trying to complicate things. Calculating your percentage return over a year from your bank balance is not very complicated (not in my case). I would estimate it as being a minute's work tops. There's no real effort or overanalysing involved. As mentioned before i agree with you on ROI.

My only dispute with you was when you mentioned that calculating your percentage profit from your bank was meaningless statistic, although i'm sure in your case it is, but not for some.

Actually now i come to think of it I don't think that asking the question "am i earning more than i did in my job?" is a simpler question as there are many other factors to consider rather than just money if changing from employment to BF. Anyway apologies as it now seems that i am becoming a tad argumentative when in essence i agree with the fundamental point you are making.

I think we pretty much agree overall with each other on this one.

I think so too. [;)]