Greening Up in Play DOUBLED MY LIABILITY!?!

[Crazyskier]

I too have noticed that the Green-up function works in mysterious ways. I'll happily use it if I have traded only one horse (team) but in other situations expect that it will give me a sub-optimal return, or on occasion a position that is worse than my current minumum result.

The attached file shows a three horse race (Test match) where the draw is also a runner. It shows that while my worst outcome is to lose £29.30, Green-up would cheerfully exchange that for a loss of £40.65 against each runner. That's the outcome it offers - not the result of applying the function in a volatile market. It's a screen I copied last night - I coud provide something similar from each time I trade.

Greening up on markets with three or more runners will always have some complexity, especially in running. I wonder, Peter Webb, whether it might though be possible to add a very simple alternative Green-up system for two horse markets, that the user culd select as an option. It wouuld involve (no doubt with a few complications) a simple single trade on the favourite. No doubt this would sometimes provide unexpected or inferior results, but it would make life easier for me and I would value it.

I'm happy to discuss of course, if this is feasible. If it exists already I'm delighted and please would you point me to it!

Hear Hear! I'm glad to see that it's not just me having these issues. There have been several instances of a single-click green up that have increased the liability beyond the worst-case scenario without. As someone who has tried several automation software products over hundreds of hours now, I should say that BetAngel is the ONLY one ever to have increased liability, and therefore I would also enjoy some modification or option to prevent this as evidently a flaw in the software.

The image above clearly shows a much worse case scenario with green-up than without it - this can't be Mr. Webb's intention?

[Tanden]

There is nothing odd or wrong with the example given.

It is a shame you haven't posted the full screenshot so we can see the current prices.

I would advise anyone who is confused by the example to calculate themselves, by hand, what they think the result would be if all trades are closed right now at the current prices.

Obviously we would need to see what those prices are.

If you do this correctly you will get the same answer as Betangel has displayed.
And the same answer that ALL other software would give - and, in this case, exactly the same result as Betfair's cash out option.

Your net stakes are in the hundreds - why would you be surprised that you could have a liability of £40?

To only lose the £29 shown if New Zealand win requires you to win enough to cover the £982 + £100 net stakes (less £29) on the other 2 selections.

So you have to win about £1053 on the New Zealand win just to make sure you only lose £29 overall.

If you close the trades now (green up) you will be removing that possibility.

This is what I warned about earlier on in this thread - when you have dutched (which you effectively have) - be careful of closing a trade on a selection that then goes on to win because you will lose the benefit of that win.

Also, please note that this is a different scenario to that of the OP.

[Crazyskier]

[quote="Tanden".
If you do this correctly you will get the same answer as Betangel has displayed.
And the same answer that ALL other software would give - and, in this case, exactly the same result as Betfair's cash out option.

Your net stakes are in the hundreds - why would you be surprised that you could have a liability of £40?.[/quote]

You're wrong. The stake is irrelevant! It's about MAXIMUM LIABILITY

If the worst case scenario is a loss of £29 - clearly shown above, then why on earth would you want an option that guarantees a LOSS OF 40!!!

The idea of green-up is to guarantee a profit OR minimise a loss, not COST YOU MORE, regardless of outcome!

You're also wrong that other software does this too - none of the many I've tried, or Betfair's own cash-out in play EVER INCREASE YOUR LOSS LIABILITY - EVER. It's BA that has this peculiar problem.

[The fish]

Hi Tanden,

Pehaps I should clarify my 'green-up' needs as a trader. I am not aiming to achieve as level a position across the runners as possible - instead I am wanting to make my worst possible outcome as good as possible. 'Green-up' here is offering to exchange a worst possible result of losing £29.30 with a guaranteed loss of £40.65. Current prices are irrelevant - there is no reason for me to take that option. (Whether it's 'odd or wrong' I leave up to you - my point is that it's daft.)

I have no need of a 'green-up' function that looks at my net stakes. I purely wish it to look at my current position and at the odds on offer to provide me with the 'best' improved position. If no combination of trades can improve the current position I would like it to sit on its hands. In fact in the example shown, two simple trades enabled me to level at a glorious -£23 across the results, though I am not expecting that an algorithm would necessarily identify that.

What I am asking Peter for is an option to use a alternative 'green-up' function that would make incremental improvements to my position, potentially applicable only (or particularly) to two-horse races. This would be valuable to me as a Bet Angel user.

[megarain]

The complexity of greening-up 2 runner or 3 runner mkts, seems absolute, when u mix trading on 2 or 3 selections.

If u just back. selection 1, the green-up is normally right.

But, if u say Back selection 1, back selection 2, and lay selection 3 ... then later, lay selection 1 .. the green-up seems to concern itself with 1 selection only .. and is normally wrong.

This phenomenon, is true with other software vendors.

Its obviously a v hard calculation.

Once again .. I understand some of the issues. Maybe BA was designed for horse-racing, with maybe .. 20 bets a race.

If we are now trying to force it to cater for literally hundreds of bets, across 3 selections .. it shouldnt really be a surprise, some odd things can happen.

I need to use BA on a live ckt mkt .. to experiment, will do this, soon.

[Tanden]

Hi The Fish

I get where you're coming from.

I just wanted to make sure people realised that your situation was different from the OP.

And that the current green up feature does work as intended but isn't appropriate in the scenario you described.

This doesn't mean it is wrong or should be changed.

As you have stated, what may well be useful is an additional button to perform the calculations you have said were possible - I currently use my own spreadsheet to do this, sometimes working across both exchanges if one of them goes down so I do know where you're coming from.

[Tanden]

Hi Crazyskier

You stated "The idea of green-up is to guarantee a profit OR minimise a loss, not COST YOU MORE, regardless of outcome!"

I don't believe that having an "idea" of how green-up should work serves us - what is important is to understand how the current feature does work and to use it when appropriate.

Obviously, if your opinion is that it doesn't work as intended then it is important to point that out as we all want useful, bug free software.

However, the current green up does exactly what it is intended to do - it closes all open trades at current prices.

It is supposed to do this and it is an essential feature of the software. It is exactly what we want it to do in many scenarios. Altering it would affect those scenarios.

However, in some scenarios it is not always appropriate and will result in a larger loss than if left alone.

This is usually when dutching has taken place and in these cases you should simply not use the green up feature.

I will admit that there is a case for disabling the green up button if the software can see that the resulting P&L would be worse than the current worst liability.

However, I hope you understand that that would not have helped you in the scenario you described.

-----------------------------------------------------------------------------------------------

You also stated "You're also wrong that other software does this too - none of the many I've tried, or Betfair's own cash-out in play EVER INCREASE YOUR LOSS LIABILITY - EVER. It's BA that has this peculiar problem."

We will have to agree to disagree on this point. I have accidentally greened up twice during an in-play delay on other software quite a few times although I did it on a single runner directly on the ladder. The worst part was I could see what was going to happen after I pressed it a second time and couldn't do anything!

But it is important to understand which scenario we are discussing.

In the screenshot that The Fish posted all other software and Betfair's cash out would give exactly the same result as Betangel.

It is simply the mathematical outcome of closing all open trades at current prices.

If you do this by hand you will get the same answer.

As stated above, it may be useful for the software to disable the green up in these situations but, again, that wouldn't have helped in your case.

It is really important for you to see that in your original post the green up feature had no way of knowing that it would worsen your situation because the green up bets that you had already submitted hadn't hit the market because of the in-play delay - so when you pressed a second time the green up feature made its calculations based on the same position as before.

But there is a possible enhancement that could help in the situation you described which I have put at the end of this post.

-------------------------------------------------------------------------------------------------

What is certain is that the software absolutely must have an option that closes and greens (or reds) all open trades at the current prices.

This option exists - it is the current green up option.

As traders it is our responsibility to understand how the available features work and when they are suitable.

In the screenshot provided by The Fish the current green up option is not suitable - but that doesn't make it wrong or mean it should be altered.

The original post is a different scenario entirely and correct use of the existing green up feature in that scenario would not have resulted in a worsened situation.

Most contributors are speculating that the green up option was clicked several times before the first bets had cleared the in play delay - if this is the case then that doesn't mean the green up option is wrong or should be changed - we need it in its current form for many other scenarios.

To Crazyskier: I don't believe the current green up needs to change but I do believe that what may have helped in your situation and so may be a useful additional feature is for Betangel to detect when a green up command has been issued and there is an active in-play delay so that in these cases the software will disable further green up commands until the in-play delay has passed.

I don't even know if that is possible and would welcome Betangel's and other members' comments.

[Dallas]

The confusion comes from how people interperate the word "Green up" this is just a terminology and Green up does not mean guaranteed profit or even minimal loss
Forget the actual word "Green up" and think of it as "Hedge up" or "Create equal P/L across all selections".

The 'Hedging up' figure shows what your net position in the market would be if you chose to close all your open positions at that moment at the best current prices to achive a equal P/L on all selections.

Weather this improves your current position depends on what money is avalible on each selection and at what odds at that given moment, as money arrives and is taken out of the market this figure is always changing and recalculating but the desision on weather to take the amount shown is always down to you - Its just a option you have and is the same option offered by all other software vendors.

If I trade on the match odds on a football market and a goal is scored i know this will put me in a good profitable position but for a few seconds after the market has unsuspeded the only money avalible to "Hedge up" against is at rediculas odds and doing it now will result in a much larger equal loss then the position i was in before the goal - if i wait a few seconds till the market begins to reform and for money to apper at better odds on each selection i can then "Hedge up" for a decent profit.

Despite nothing actually being traded on any selection just money has now been offered in the market at better odds for me to hedge up against.

[Crazyskier]

Tanden, thanks for taking the time to elaborate why you think that Fish's -£29 figure SHOULD become a -£40 figure and also how our issues and double clicking the green-up can adversely affect liability. Thanks also to Dallas for clarifying how green-up can mean lay liability equally across all selections or red-up, etc.

However... to refine this example given by Fish and propose a suggested fix I'd simply say this:

When greening/redding up the algorithm should SURELY just take the green figures and LAY THEM ONLY, whilst BACKING the current red ones? This IS INDEED WHAT BETFAIR does, and also the other programmes I've used but will not name here (as they get edited out anyway).

What BA is doing is NOT what you can do manually as Fish explains - he made the -£29 outcome reduce to -£23 on ALL selections, just by LAYING THE GREEN selection. This is both logical and should be how BA works ideally, regardless of dutching or the number of outcomes. You Tanden are absolutely incorrect to suggest that making a worst case scenario of a £29 loss result in a £40 loss is ever how BA SHOULD work correctly. It simply is NOT. And as I've stated many times Betfair's cash-in does the same, simply laying greens to spread the risk across all selections, NEVER increasing the risk amount - EVER.

[Tanden]

To Crazyskier

Firstly, I apologise for yet another long-winded post! I seem to be incapable of being concise.

Secondly, you stated in your last post...

"You Tanden are absolutely incorrect to suggest that making a worst case scenario of a £29 loss result in a £40 loss is ever how BA SHOULD work correctly."

But I did not state that BA should turn a £29 loss into a £40 loss.

I have stated that the current green-up function is correct for many scenarios and therefore needs to stay as it is. If you can see it would create a worse situation, as in the example given by The Fish, then don't press it.

For those specific scenarios I support the suggestion of having an additional green-up function which should work in a similar manner to how you have described in your last post but it is vital that it is in addition to the current green-up and does not replace it.

I have a spreadsheet which does these calculations so know that they are possible and I have duplicated The Fish's example and my spreadsheet achieved the equalised £23 loss that The Fish spoke of.

As far as other software being able to do this I am going to completely concede that point to you. I don't know of any but that is only because of a limit to my knowledge.

Nor am I very familiar with the workings of Betfair's cash out feature, so again I will concede that point also. Although I have seen many situations where I was better off greening-up via the software than using Betfair's cash out. So both have their value.

Getting back to your original post I am under the impression that you believe the current green-up function caused your liability to increase from £50 to about £98.

I still maintain that this is not true and that the reason your liability increased was because you fired multiple bets into the market whilst the in-play delay was effective - and other software does this too - I have made the same mistake myself using it.

Had you only pressed the button once then your liability would have reduced.

No matter what method of greening is used if you fire in multiple bets by mistake you are likely to end up in a worse situation.

Whilst in a previous post I suggested that there could be merit in having BA disable multiple green-ups I am not convinced that this is the best use of the development team's time when all a trader has to do is wait for the in-play delay before trying to green-up again.

So, to summarise, I totally agree with having an additional green-up function (which looks to lay the selections with higher P&Ls and/or back the selections with lower P&Ls) but you should also understand that this is an extremely risky way of equalising the P&Ls in a fast moving in-play market especially when there are selections with no bets matched.

And, ironically, it wouldn't have solved the situation described in your original post.

But what is it that you think the software should have done in your case?

You said you were dutching to a liability of £50 so perhaps you had a few runners showing an equal profit and several/many un-dutched runners showing that liability of £50.

There are only really two approaches to equalising the profits - either lay the few dutched runners (which is what the current green-up does) or back all of the others.

Which one gives the best result will depend on the current prices so you can't dismiss the current green-up function as wrong just because there are times when it isn't appropriate. It will often give the best result.

It is also much safer in certain cases because as each of the submitted bets gets matched the liability reduces so you aren't left with a larger liability if they are not all matched.

But, if you were to have the software take the second approach and back all of the un-dutched runners then as each of the submitted bets gets matched the liability on the remaining runners increases and will only decrease once every single bet is matched - in your example of the closing stages of a horse racing market this is extremely risky. What if you matched all but one of your new back bets and that last unmatched bet left you with a runner carrying all the increased liabilities and it went on to win?

This is why I support an additional green-up function whilst keeping the current one and keeping them as separate so that the trader can decide which to use depending on market conditions.

[The fish]

I believe Peter Webb monitors the forums - I think he made an earlier contribution to this one. Given the consensus that appears to have developed for additional or extended green-up capability I would be interested to hear his thoughts on whether he would support that being included in the development plan for Bet Angel. I would certainly value it and I suspect, given this discussion so far, that many others would too.

[The fish]

Am I wrong to think that the forums are monitored? Is there an alternative way of raising a request to enhance the Bet Angel product? Can anyone advise?

[Dallas]

Am I wrong to think that the forums are monitored? Is there an alternative way of raising a request to enhance the Bet Angel product? Can anyone advise?

Yes the forum is monitored but with so many threads on so many different topics there is no guarantee any suggestions buried in these particular threads are seen and more importantly no one can add to them
There is a section especially for users to put suggestions forward which i posted the link to earlier in this thread but here it is again and this is monitored by the development team

viewforum.php?f=20

[The fish]

Thanks, Dallas.

I've put forward a suggestion for enhanced green-up as you propose, aligned with the consensus in this forum.

Here is the link, viewtopic.php?f=20&t=11470&p=97488&hili ... +up#p97488, for any who want to read it and vote in the survey that assesses demand.

Vote now to achieve green-up change!

[Crazyskier]

The yes vote just says 'form invalid' so I seem unable to vote on that linked page.