I always had the impression that Optimark was hurt by two problems...
As with any complex state of affairs there were many. My personal opinion - one widely shared, one uniquely my own - on the issue is this: first, and commonly recognized in the securities industry, is the fact that OptiMark sought to put portfolio managers in the sellside trader's seat, by allowing them to execute their own trades. That approach neglects every aspect of the buyside-sellside dynamic and all of the elements, including but not limited to the relationship factor, that make it work so well.
The second reason for OptiMark's demise - again, among many - in my opinion, is not so much a securities industry factor but a general business factor, as real in institutional brokerage as it would be in the car industry, pharmaceuticals, or any other: a product was created without a need for it; without there having been a niche to be filled.
And, yes, the profiling system wasn't user friendly, it had no liquidity, and by situating itself in the middle of continuous market hours, it went head-to-head from day one with the inextricably entrenched dealers, brokers, and alternative trading systems already there.
of course getting a fill on a $20 stock at noon for $17 is better than a VWAP fill on the same $20 stock for $19.75. I assume the example is hyperbole -- the $20 stock does not likely sell for $17 at noon and still have a VWAP of 19.75.
Assume nothing. No hyperbole whatsoever.
The stock need not hit $17, let alone print a single $18, for a block trader to print $17 for his/her customer. And that is the exact point: with capital committment, one can easily undercut a VWAP, RPM, or other brokers' bid for a block.
If your assumption is that every time a trader works a block actively, he or she achieves a better price, well, of course that would be better than VWAP if your assumtpion were correct.
Where have I "assumed" anything; let alone something as ridiculously unconditional ("every time") as you're suggesting in this statement?
And interestingly, I find your resistence to saying that the VWAP price might ever represent price improvement to what a good active trader might achieve as reflective of one of the major obstacles that this particular system faces -- convincing market participants (such as yourself) that a different (and worse still, an electronic) way of doing this might on some occasions actually be advantageous to what a toptrader can achieve, and thus to use it when it makes sense to do so.
Look. You've asked me for my experience and my opinions, and you're getting it.
Do I think that this ATS might be an improvement over what desk traders can do?
Well, sure. Sometimes. My point is, though - if you guys understand what I mean by this: there's no reason it has to improve desk trades. Ever. As long as traders can, will, and do price blocks below or above the market, he or she can beat any methodology.
For a stock that opens at $50, trades up to $53 intraday, and closes at $51, the VWAP might be $52.50, $51.75, or $50.90, depending upon where the bulk of shares were traded. Lovely. Any trader at a desk can outbid that if they want the stock, if they want the business. If the number at the end of the day wound up being $51.75, he can print a customer at $48, $49, $50, $51 - whatever his experience, capital committment resources, analytics, and personal estimation of the nature of the business relationship justify.
In fact - along the lines of a practice I'll bet we'll see emerge in the continuous market - I'll bet that buysiders could find more use in the system for pricing leverage with human trading desks than in the actual use of the ATS.
Think about it: a portfolio manager can call a sellside desk and say, "The VWAP today was $51.75, and you know, there's this new ATS...but I've done business with you for 10 years, so, what can you do for me?" And the trader will simply say, "OK, how does this sound: from now on, on orders you want VWAPed, at the end of the day we'll take whatever the price was you would've gotten from that algorithm, and I'll print you 200 basis points better, buy or sell."
When you see prints going by under or over the market, that's exactly what is going on. Block distribution away from the market, almost always at a price that will prove superior to whatever the day's VWAP is.
Of course, not all trades are big enough - nor are all customers worthy enough, truth be told - to justify printing above/below market prices or giving huge haircuts on blocks.
I hope we are not coming across as too strident or zealous
Hehehe. Almost, but not yet. :) I'm happy to continue this dialogue, answer other questions, etc.
LPS5 |
