Weight of Money

A place to discuss anything.
Post Reply
Archery1969
Posts: 4478
Joined: Thu Oct 24, 2019 8:25 am

Hi,

So, from what i have read trading by Weight of Money (WOM) use to work well many years ago when opening, cancelling and closing trades.

But over the years the markets evolved, people and bots got clever and started placing large sums of money at either end of a selection to push a price one way or the other and usually forcing others to take a loss. Also, some could argue that the calcualtion for the WOM is very sensitive to slight increases or decreases in amounts on offer waiting to be matched.

It seems the general concensus now is that it cant be used anymore due to:

1. Spoofers
2. Sensitivity

Can we address the above points to make it work in our favour again ? :roll: :roll:

Lets take an example:

Inputs

Back Volume:
B1 = Best Back Amount (£100)
B2 = 2nd Best Back Amount (£50)
B3 = 3rd Best Back Amount (£50)

Lay Volume:
L1 = Best Lay Amount (£200)
L2 = 2nd Best Lay Amount (£50)
L3 = 3rd Best Lay Amount (£800)

WOM 1 = (B1 + B2 + B3) / (B1 + B2 + B3 + L1 + L2 + L3) = 0.16%

Allowing for Spoofers

One suggestion is to weight the amounts on either side, giving more emphasis to money waiting at the front apposed to money waiting towards the back. To get to 100% I have used 56% for money at the front, 33% for money in the middle and 11% for money at the back.

WOM 2 = ((B1 * 56%) + (B2 * 33%) + (B3 * 11%)) / ((B1 * 56%) + (B2 * 33%) + (B3 * 11%) + (L1 * 56%) + (L2 * 33%) + (L3 * 11%)) = 0.26%

Allowing for Sensitivity

One suggestion is to calculate WOM in a different way to make it less sensitive to relatively small changes in the amounts on either side. A calculation I found on other website was someone who wrote a pricing model over 7 articles using the following calculation.

ABS((Back_Volume - Lay_Volume) / (Back_Volume + Lay_Volume))

So, using the new method and also incorporating the spoofers percentages we get.

WOM 3 = ABS(((B1 * 56%) + (B2 * 33%) + (B3 * 11%) - (L1 *56%) + (L2 * 33%) + (L3 * 11%)) / ((B1 * 56%) + (B2 * 33%) + (B3 * 11%) + (L1 *56%) + (L2 * 33%) + (L3 * 11%))) = 0.47%

Most information suggests we enter a back trade if WOM < 33% and a lay trade if WOM > 66%.

To recap:

WOM 1 = 0.16%
WOM 2 = 0.26%
WOM 3 = 0.47%

As you can see from above, if using WOM 1 we are open to spoofers, if using WOM 2 then its better but the ratio is still probably not enough to put real pressure on prices one way or the other. By using WOM 3 we would both deal with spoofers and mean getting to < 33% or > 66% would require significant money on either side.

Obviously there are still no guarantees as money can still get pulled or matched in an instant especially in low liquidity markets but i think WOM 3 is a great deal better when deciding to open, cancel or close trades.

Anyway, hopefully i have explained the above ok. Anyone got any thoughts, even if you consider it total b*llocks. Be kind though. :)
User avatar
Dallas
Posts: 23571
Joined: Sun Aug 09, 2015 10:57 pm

The option to weight the amounts is already there, you can find it in the 'Display' tab of your main settings area
https://www.betangel.com/user-guide/display.html

Also have you looked at using the Stored Values and 'Money Waiting' options?
You could do almost anything you want with these, there's a servant example here to get you started
viewtopic.php?f=58&t=18645
elofan0
Posts: 415
Joined: Fri Jan 13, 2017 4:44 pm

I think no matter what WOM option is used you are still relying on transmition of data if for instance there is a slight hold up on the money being processed ... kind of like refresh rates if you had it set say 500ms the data reaching would not be the same as set at 200ms but i know what your saying
CallumPerry
Posts: 575
Joined: Wed Apr 19, 2017 5:12 pm

Thinking logically, the low hanging fruit got picked years ago so WOM naturally would become less effective. A few years ago I tried a clever seesaw calculation that look the other runner's volume into consideration too no matter how many runners were in the race. It was my first attempt at getting good with Excel formulae and from there, diving into VBA.

You can add loads to get a detailed indicator but the problem I found was that it was always lagging! You have to do something clever with the WOM/evolved WOM by adding other bits to it. Some traders may still use WOM calculations in their systems but the profitable ones probably have a few more readings as part of their entry/exit criteria.
Emmson
Posts: 3577
Joined: Mon Feb 29, 2016 6:47 pm

Atho55
Posts: 678
Joined: Tue Oct 06, 2015 1:37 pm

Archery, thanks for doing the analysis. I`ve adjusted my settings as a test.

I find that converting the WOM value to a 1 or 0 via SV makes a good trigger. My test rule looks for a WOM% or 1 on Rank 1 to Back AND WOM% on Ranks 2-6 to Lay such that both must equal 1 at the same time. The idea being that movement on both could happen to maintain the Book%.

What I have noticed is that the trigger on Ranks 2-6 very often triggers more than 1 runner suggesting that the WOM% has a spread. Has anyone noticed similar?

The 1/0 looks like this. The RHS columns


WOM.jpg
You do not have the required permissions to view the files attached to this post.
rik
Posts: 1583
Joined: Sat Jan 25, 2014 5:16 am

amount and total volume of back bets placed relative to lay bets by far most important indicator, forget about WOM
Atho55
Posts: 678
Joined: Tue Oct 06, 2015 1:37 pm

Rik, are you able to point to what would be a candidate in the remaining races
Archery1969
Posts: 4478
Joined: Thu Oct 24, 2019 8:25 am

rik wrote:
Mon Jan 18, 2021 5:37 pm
amount and total volume of back bets placed relative to lay bets by far most important indicator, forget about WOM
Rik,

1. Do you mean volume of back amounts and lay amounts already matched.
2. Is there an indicator for such.
3. How would you automate.

Cheers,
Post Reply

Return to “General discussion”