Looking for slippage data, i.e. quote was "X bid, Y ask" when I sent the order, average execution was "Z".
Thanks.
Thanks.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Anyone have any performance data on IB's Arrival Price
- Thread starter ericholtman
- Start date
Have you looked at their website? Maybe here
http://individuals.interactivebrokers.com/en/p.php?f=smartRouting
http://individuals.interactivebrokers.com/en/p.php?f=smartRouting
Oh, I've seen the marketing literature.
I was wondering if anyone had any actual trading diagnostics they were willing to share.
I was wondering if anyone had any actual trading diagnostics they were willing to share.
Got it. I've long been interested in tracking the same for myself but it's not as straightforward as you'd like it to be. I'm frankly surprised that there's not a small lightweight piece of client software that could "listen" for trade messages and record the current bid/ask/mid condition along with trade price. I posted here a year or so back asking if others would be interested in such a product and heard crickets chirping
In the end I've defaulted to benchmarking my trades (ex-post) against end of day closing price, which you can get from the flex reports. Admittedly is far from perfect, but over huge numbers of trades I think it gives an idea of fill quality.
In the end I've defaulted to benchmarking my trades (ex-post) against end of day closing price, which you can get from the flex reports. Admittedly is far from perfect, but over huge numbers of trades I think it gives an idea of fill quality.
To allow meaningful comparisons, aren't you also going to need (per symbol/instrument, and across all venues):Quote from ericholtman:
Looking for slippage data, i.e. quote was "X bid, Y ask" when I sent the order, average execution was "Z".
Thanks.
- average volume typically filled per some unit of time
- # orders typically resting at X - (n * minimum tick size) for n = 0 to ...
- # orders typically resting at Y + (n * minimum tick size) for n = 0 to ...
...?
Quote from abattia:
To allow meaningful comparisons, aren't you also going to need (per symbol/instrument, and across all venues):
Not particularly.
I already track skid in my trading system. So I know, if I place a market order, what the quote was going in, and what the VWAP is on the executions.
If I switch over to Arrival Price (or any other algo), I'd compute the skid the same way.
I'm not interested in "well, the book was X deep, the algo should have known to get off Y shares there". I'm interested in "for the trading I do, is algo A better than algo B".