The diff is the price the diff order would be submitted at.The guy was asking how it works with ninja.SLM would be submitted at the specific price entered,once the level is reached,whereas STP would be submitted at market once the level is reached.
For futures I use an auto-improve order so I will not cross the market and on equities I use discretionary orders, here again I don't have to worry about crossing the market. My order will not be sent until 2 NBBO ticks print and also I have a percentage alert that will warn me if the NBBO has a wild swing. It is all auto - I hit one button and done.
I have played around with using stop limit offset orders, and to me they made no sense. Why would you want to set a buy stop limit of, say 50.00 in CL, with an offset of +5? So when 50.0 is reached, the stop triggers the limit order to buy at 50.05. Well, why would one want to get into the market 5 ticks later than where they think the proper entry is? Likewise, if CL is at 50.00, and I set my buy stop limit order there, why would I set it to -5, thus putting my buy limit order to buy at 49.95, 5 ticks below where I think the proper entry is? Why would I just not place a buy limit order at 49.95? Stop limit orders, in my experience, add a layer of confusion and complexity and uncertainty to a business that has enough of that already.
fordewind, I think you and I are talking about different peas in the same pod. I know what a stop market order and a stop limit order is. My whole point in this is what was the definition of a "marketable limit order", because I had never heard that term before? I thought maybe it was some new funky way of trading. But it's not. The question was answered by the poster of the question in his inference of limit orders "becoming" market orders. That's bollocks. Neat trick that would have been!
For the second example, I would imagine so that your order isn't visible until just before it would fill. The other reason for buy stop limit orders is so that you can buy, when price is below you, but limit it to your specific price. If price is at 50 as you say, any prices above are limit prices to sell, and any prices below, are limit prices to buy. But if you say want to sell when price goes down lower, it needs to first get there, tick below the price, and then your limit order can be placed above the price. This would I imagine be the only way to have an order to sell at a specific limit price when price is still above where you want to sell at.
I'll give you a perfectly valid reason: unless your platform handles submitting "stop protect" orders then no "stop market" order you put out there will actually sit on the exchange. For instance, CME supports stop-limit and stop-protect orders, it does not support stop-market orders. So when your platform submits an OCO bracket with a stop market order, that stop has to go somewhere once your main order is filled and in a lot of cases it sits on your FCM or trading network's servers (if not local - which you sure as hell do not want). It's only when the FCM/trading network notices that price has hit your stop that a market order is submitted to the exchange which means you're just about guaranteed to be near the end of the queue. Now if one uses a stop limit order the stop sits *on the exchange* and not on any in-between FCM/provider/etc meaning that the limit is submitted to the book with as little latency as possible. Try taking 40 ticks of slippage in gold or crude sometime to know why this is a good thing. Another completely valid use case is when one is dealing with instruments that have wide spreads (e.g. 4-10 ticks) and the use of the positive offset limit is effectively a "pay-up but don't totally screw me" range. Invalid order and will be rejected by the exchange. It's actually a shame because one can use such orders as triggering type entries the stop price is basically a trigger to throw a limit order out there such that it will be filled on a retrace (think -20 or -30 ticks, not -5). Why are we even talking about stops anyway when the original issue was about marketable limit orders?
Because the question was what is a "marketable limit order", and the discussion fell apart from there.
oh I'll just give it to, you win. I was trying to explain to the guy who was asking about stop limits in black swans. A sell stop limit would be a sell limit with a limit way down there. Like a market order on the way down and if the stop trades through then just like a normal limit waiting for a better price.
I think worrying about HFTs in general is pretty paranoid. Aren't there HFTs designed to spoof limit orders to fool the other HFTs? If an HFT's strategy is so simple to see, wouldn't someone build an HFT to take advantage of the other HFT's strategy? No HFT firm would be content to let a rival take a disproportionate slice of the pie.