What's a decent Profit / Turnover figure on a strategy?

[Euler]

I measure profit / turnover to give me an idea of if my margins are shrinking or expanding. They have been pretty stables for years, but it's the market that is the key influencer. If I can put through decent sums I obviously do better. But if the market is think I can't push through as many orders and I do worse at the top line. But margins seem to be static for quite a few years.

I tend to invent new strategies to grow my over all pool of profits, but they are all measure separately as they all have different metrics.

Ultimately you can break down any profit into an equation to see if you can tweak part of that equation to improve yield. So it's a worthwhile exercise.

[spreadbetting]

I don't understand why you would use your exposure as a measure of turnover, makes comparing roi of strategies that back/lay only impossible. Profit over size staked/settled is what I have always used.

Why would using opening exposure make comparing lay or back strategies impossible, surely laying a 10/1 shot for £100 exposure is no different than backing £100 @ 1.1. You're simply comparing a bet's stakes and odds regardless of which way round you do it.

[LinusP]

I don't understand why you would use your exposure as a measure of turnover, makes comparing roi of strategies that back/lay only impossible. Profit over size staked/settled is what I have always used.

Why would using opening exposure make comparing lay or back strategies impossible, surely laying a 10/1 shot for £100 exposure is no different than backing £100 @ 1.1. You're simply comparing a bet's stakes and odds regardless of which way round you do it.

Like Peter I use roi to compare strategies over time and against each other, its especially helpful when increasing stakes and A/B testing.

For example see graph below for two strategies, one only lays and the other just backs, using profit/size settled the roi can be compared, using exposure they can't. The profit itself factors in your 'exposure' when you win/lose so it doesn't make sense to calculate them differently.

roi = profit/size_settled

Screenshot 2019-04-18 at 15.22.02.png

roi = profit/exposure

Screenshot 2019-04-18 at 15.30.54.png

As above but zoomed in

Screenshot 2019-04-18 at 15.32.15.png

[spreadbetting]

I've no idea what size_settled is? Is it simply the stake returned from a settled market i.e. a £100 lay at 10 your settled stake is £10 win or lose?

I can see the point using stakes as a comparision measurement and use it myself to check I'm gaining value of both my backs and lays by simply looking at the agregate pnl of lays and backs separately. BUt to call it a return on investment is the only issue I'd have with it as I'd always view laying a 100's shot and backing a 1.01 as the same thing if I want some guage of my return to turnover.

But at the end of day as long as people are consistent in their approach they'll be able to guage whether they're progressing or not. For me these days I'm going backwards profit per market wise but overall yearly pnl is up simply because I churn over more markets.

[ShaunWhite]

Answer still seems to be to calculate whatever you want for your own use, but it's pointless comparing your apples with other people's oranges.

The factor nobody has mentioned is time, all measures of performance require it. if I made £100 by turning over 10grand in an hour then that's better than if i'd previously taken all day to do it. Both have the same return but the former has earned me 7 hrs to do whatever I want, trading or otherwise. Where's the value of that in the calcs?

[LinusP]

Answer still seems to be to calculate whatever you want for your own use, but it's pointless comparing your apples with other people's oranges.

The factor nobody has mentioned is time, all measures of performance require it. if I made £100 by turning over 10grand in an hour then that's better than if i'd previously taken all day to do it. Both have the same return but the former has earned me 7 hrs to do whatever I want, trading or otherwise. Where's the value of that in the calcs?

Good point, when comparing, my comparisons are always over the same time period for example a month.

My point in regards to roi is that if you calculate per liability you are comparing your own apple and oranges but I will give up making that point now.

[Derek27]

If you trade £3000 on a market, there's a big difference risk-wise, whether it was one £3000 trade or 6 X £500 trades with a green-up between each trade, but the raw turnover is the same. That's quite difficult to factor into any performance figure without a lot of tedious record keeping.

[LinusP]

If you trade £3000 on a market, there's a big difference risk-wise, whether it was one £3000 trade or 6 X £500 trades with a green-up between each trade, but the raw turnover is the same. That's quite difficult to factor into any performance figure without a lot of tedious record keeping.

I am not sure where risk has come into this as ROI is all about measuring return or ‘performance’.

I agree that a 3k outright bet is very different to 3k trading but you would never compare the two, this is all about comparing similar strategies or measuring the change in performance for a single strategy over time.

[Derek27]

If you trade £3000 on a market, there's a big difference risk-wise, whether it was one £3000 trade or 6 X £500 trades with a green-up between each trade, but the raw turnover is the same. That's quite difficult to factor into any performance figure without a lot of tedious record keeping.

I am not sure where risk has come into this as ROI is all about measuring return or ‘performance’.

I agree that a 3k outright bet is very different to 3k trading but you would never compare the two, this is all about comparing similar strategies or measuring the change in performance for a single strategy over time.

The 'I' stands for investment, which implies money put at risk. It's possible to put £1K in an inactive market where you could easily scratch the trade without risking much, so trading turnover can be quite misleading.

For evaluating performance, profit on capital allocated to a particular strategy could be a better measure.

[LinusP]

If you trade £3000 on a market, there's a big difference risk-wise, whether it was one £3000 trade or 6 X £500 trades with a green-up between each trade, but the raw turnover is the same. That's quite difficult to factor into any performance figure without a lot of tedious record keeping.

I am not sure where risk has come into this as ROI is all about measuring return or ‘performance’.

I agree that a 3k outright bet is very different to 3k trading but you would never compare the two, this is all about comparing similar strategies or measuring the change in performance for a single strategy over time.

The 'I' stands for investment, which implies money put at risk. It's possible to put £1K in an inactive market where you could easily scratch the trade without risking much, so trading turnover can be quite misleading.

For evaluating performance, profit on capital allocated to a particular strategy could be a better measure.

How do you measure your performance over time?

[Derek27]

Answer still seems to be to calculate whatever you want for your own use, but it's pointless comparing your apples with other people's oranges.

The factor nobody has mentioned is time, all measures of performance require it. if I made £100 by turning over 10grand in an hour then that's better than if i'd previously taken all day to do it. Both have the same return but the former has earned me 7 hrs to do whatever I want, trading or otherwise. Where's the value of that in the calcs?

I have a spreadsheet that calculates my daily compounded and monthly profit on capital. It makes great reading if I make a few percent on New Year's day and see how many £Bs I'm projected to make for the year and I can enjoy a pipe dream for a few days.

[Derek27]

I am not sure where risk has come into this as ROI is all about measuring return or ‘performance’.

I agree that a 3k outright bet is very different to 3k trading but you would never compare the two, this is all about comparing similar strategies or measuring the change in performance for a single strategy over time.

The 'I' stands for investment, which implies money put at risk. It's possible to put £1K in an inactive market where you could easily scratch the trade without risking much, so trading turnover can be quite misleading.

For evaluating performance, profit on capital allocated to a particular strategy could be a better measure.

How do you measure your performance over time?

Raw profit is the best guide. There is no really accurate way as even comparing profits or profit percentages for the same month or quarter last year can be affected by abandonments due to weather, changing markets, reductions in market turnover, etc.

[ruthlessimon]

All about balance between the big 3:

1. Edge (Profit/Turnover)
2. Frequency
3. Scale

Edge * Frequency * Scale = Profit

Now it's tricky to improve an edge, & it doesn't necessarily mean it's worth the pain; here's why

I'm gonna make my entry criteria harder to trigger (1 = easy to trigger, 12 = hard to trigger).

Notice how I can choose to "improve my edge" - at the cost of 50% profitability. Worth it?

Further to that I've sacrificed half of all my entries (trade frequency is down 50%); this will be tricky to handle mentally, & also brings up questions as to whether the sample is even big enough to be valid.