Same ECN, different fills: how last look changes execution

[Shadow Trader]

A broker advertising "ECN" or "STP" execution tells very little about how orders are actually treated one layer deeper. A big factor is last look.

Last look is a practice where a liquidity provider can take a short window after receiving an order to decide whether to honor the quoted price. If the market moved against them in that window, the order can be rejected or re-priced. If it moved in their favor, the fill goes through.

For a swing trader holding for hours, that's mostly noise. For a scalper whose edge is a pip or two, it's a direct tax. Winning orders get rejected more often than losing ones, and nobody has to call that manipulation, it's just how the provider protects themselves from fast clients.

Some brokers pool liquidity from providers who use last look and from providers who don't. The mix changes by pair and by time of day, which is why fills can feel different at 3am than at 9am.

Asking the broker whether their liquidity providers apply last look is a perfectly fair question. Some will answer plainly.

Quotes from last-look providers also tend to look slightly better on paper than quotes from firm providers, because the provider keeps the option to decline. The tighter displayed spread is partly paid for by the rejection risk. A feed that shows 0.1 pip spreads with a 12 percent rejection rate on fast orders may be worse in practice than a feed showing 0.4 with almost no rejections.

Some brokers publish a rejection or fill ratio. Others will share it on request. A scalper who can see that ratio by hour of the day knows where the pricing is real and where it is decorative.

The simplest personal test is counting how many orders were rejected or re-priced in a week, then splitting the count by winners and losers. If rejected orders would mostly have been winners, the effect is real.

Has any broker ever given a straight answer on this, or is it all marketing language?