Hi Dallas - any chance you could help me with an existing couple of posts? - you've always been my hero in the past..... maybe it was a stupid question (if there's any such thing..) but i noticed several people asked the same question from a couple of years ago and didn't get an answer either....
From 3 days ago:
It would be really helpful to know (on any given market) if the money coming in was back or lay money. . eg: if the volume bars were either blue or pink it would be very useful. Would it be possible to effect this change? At the moment i try to refer to several indicators to try and anticipate pricing direction. Thanks
From 2 years ago:
in advanced chart settings could we have an option for lay stakes & back stakes to be coloured, i.e. have a choice for back stakes to be red & lay stakes in green?, (or any colour one chooses?), also maybe candlestick colouring for both back stakes & lay stakes?
Pretty please..... cheers
Help - unanswered posts.....
It's not possible to show on the advanced charts which side the money was offered/taken from
I will just show the amount matched which will always be made up of 50%back and 50% lay
I'm almost certain this can't be grabbed from the API directly either, the only workaround would be to calculate it locally via the money waiting and traded volumes, but that will require loads of calculations dozens of times each second and even then it won't always be 100% due to money being pulled, large bets taking out multiple prices etc. If done in excel you could probably then chart it in some way.
I think ShaunWhite has posted somewhere that he's looked down this route in more detail, perhaps if he sees this he can comment
I will just show the amount matched which will always be made up of 50%back and 50% lay
I'm almost certain this can't be grabbed from the API directly either, the only workaround would be to calculate it locally via the money waiting and traded volumes, but that will require loads of calculations dozens of times each second and even then it won't always be 100% due to money being pulled, large bets taking out multiple prices etc. If done in excel you could probably then chart it in some way.
I think ShaunWhite has posted somewhere that he's looked down this route in more detail, perhaps if he sees this he can comment
-
ShaunWhite
- Posts: 10690
- Joined: Sat Sep 03, 2016 3:42 am
You're right Dallas, it's tricky, the best I could do in Excel was to look at the LTP, see which side that price was (is?) on, and then use the selection vol change to put that money in one basket or another. The issue was conflation of the data as each 20ms market change could contained the volume change of several trades, but only had a single LTP. And of course relativistically it takes the perspective of the taker not the maker when determining if it was backing or laying money. It would work better on slow moving markets than when it gets hectic. You have to do a similar thing even when using the api directly and although better, it's still a finger in the wind rather than something 100% accurate.
It was interesting for a while but I can't say I (l)eant much from it, it's been shelved though rather than binned. I think the most interesting thing I saw was that some moves require money (ie they push against/through resistance or support), but other moves are much more the result of offers being withdrawn. I looked at #backbets vs #laybets too and if I remember correctly there were about 2.5 times the number of back bets vs number of lay bets, but that wasn't a suprise when you consider your average participant is more likely to be a backer than a layer.
So, there might be something in there or there might not, but wrangling some new numbers out of the market and looking where others don't is all part of the fun. Let's face it if you're looking at the same ideas and data as 5,000 other people you're unlikely to find any untrodden snow, just the yellow stuff.
This was the thread where it originally got a mention -> viewtopic.php?f=17&t=16116&hilit=ltp
It was interesting for a while but I can't say I (l)eant much from it, it's been shelved though rather than binned. I think the most interesting thing I saw was that some moves require money (ie they push against/through resistance or support), but other moves are much more the result of offers being withdrawn. I looked at #backbets vs #laybets too and if I remember correctly there were about 2.5 times the number of back bets vs number of lay bets, but that wasn't a suprise when you consider your average participant is more likely to be a backer than a layer.
So, there might be something in there or there might not, but wrangling some new numbers out of the market and looking where others don't is all part of the fun. Let's face it if you're looking at the same ideas and data as 5,000 other people you're unlikely to find any untrodden snow, just the yellow stuff.
This was the thread where it originally got a mention -> viewtopic.php?f=17&t=16116&hilit=ltp
Thanks Dallas & Shaun for taking the trouble to reply. 'Not sure i fully understand though, but it's helpful to get an answer from people who know what they're talking about. I didn't realise the bar charts reflected the MATCHED volume, makes perfect sense when i think about it, another little light bulb moment.... they all add up. Thanks again. & back to the drawing board.... Cheers.