RobertB

I have a slightly different take on this. On first read this may not be a popular opinion, but I think it should be presented. We all know that there are no free lunches. Only socialists and those not versed in economics think so. I would be very surprised to find very many socialists here on this board full of individualists (not to get too political here). We traders are an individualist breed by nature, and I don't assume anyone here, upon reflection, believes that the world is full of free lunches. Other than within families, free lunches do not exist. Regarding IBs fees. I prefer IB's system of various fees passed on to me when appropriate than to higher set commission rates without fees. If I place a trade that removes liquidity, for instance, and by my doing so incur a fee to IB, I have no problem with IB's passing that fee onto me. The only way IB can profitably provide rock-bottom, no minimum purchase, commissions is by passing on fees when they occur. It is annoying, I must admit, not to always know what the precise commissions will be in advance for any particular trade. But I prefer that to higher-priced, but known-in-advance, commissions offered by IB's competitors. For instance, with TOS I get a known-in-advance rate of $1.25 per contract. But IB, even with all its potential fees, will always be less than that. (However, for some, perhaps knowing in advance the precise commissions is worth the extra price; if so, that's totally fine, of course.) The bottom line is that I understand the $0.70 per contract IB commission rate to be just the base price with possible credits and debits added or subtracted. Transaction costs come in from many areas. I'm just happy that IB has been able to get its own costs down to where it can profitably offer $0.70 contracts, give or take. If IB wanted to set an invariant, fixed commission rate (no fees, no rebates, etc.) it would probably raise its commissions rate to somewhere between $1 and $1.25 per contract, which in the long run--even though you'd avoid the annoyance of not knowing fully the commissions in advance--would result in higher prices overall. Again, there are no free lunches (as we all know).

DShaver, I wouldn't be afraid to use IB. You just have to get used to its vagaries. They do provide a sandbox account for safely learning TWS's features. And it's worth learning because of IB's superior commissions and ability to fill trades. Once you learn IB's quirks, you as a trader should benefit from the rest.

What Kim says is quite true. You do get used to the negative numbers after a while. But even to this day when I trade using negative prices on IB, I'm forced to add what should be unnecessary "brain cycles" to the trade to make sure I get it right. Just remember that when you see negative values, the current market for a buy is at the (negative) bid; the current market for a sell is at the (negative) ask. As I said earlier, it is a very poor design to make users have to think about this. TOS doesn't use negative prices. Instead, they go the more natural way, dispensing with formal definitions in order to have a more clean user interface. For instance, TOS debit trades are always a buy; credit trades are always a sell. Thus, on TOS when opening a trade, you'd buy a RIC and sell an IC (the opposite of the way you'd make the same trade on IB).

So, you ran into the infamous IB negative prices. These confused me to no end when I first encountered them. And when I called IB to ask about them, I was subjected to ten minutes of classic IB Support gobbly-gook. The guy on the line had absolutely no idea what he was talking about. His explanations were beyond anyone's ability to comprehend. He babbled on about my RIC's being somehow an impure iron condor that he'd never seen before. That I was supposed to be selling the inner strikes. That the negative prices were warnings of some sort. The only real warning given by those negative prices is that many areas of IB's software interface suck. When I open a position for a credit, I think of it as "selling." When I open a position for a debit, I think of it as "buying." The following is going to sound confusing, but here goes. I believe this is how IB's software works: If you are buying for a debit, IB uses positive price values. If you are selling for a credit (both opening and closing a position), IB still uses positive prices. However, if you are buying for a credit (such as for opening an IC on IB, (which IB considers a buy even though it's a credit trade), IB still uses positive price values, but turns around and uses negative price values when selling for a debit (which is what it considers a RIC when you open a position). I know that sounds confusing, and it is. This is all because (apparently) IB considers opening an iron condor, which is a credit trade, as a "buy." This is a classic example of poor software design. Their software design requires users to deal with issues that the software itself should deal with on its own. This is similar to when you are purchasing something online. How many times have you filled out a form that asks you to remove the spaces when entering your credit card number? Why doesn't the software itself handle such a trivial task? Poor software design is very common.

For a number of years I've programmed with IB's API using C++ and Visual Basic. If you have the technical background, it's pretty easy. IB provides a lot of examples which you can easily modify to suit your purposes. However, I would urge you to think very carefully before writing any code that handles order execution. For example, when using TWS to submit an order, the order goes to IB's servers. At that point, IB is responsible for executing your order, including limit orders and stop-loss orders. If IB screws up, it's their fault and you have recourse. But, in contrast, if you write your own code to handle these sorts of things, especially stop-loss orders, these types of orders are handled entirely on your computer using your own programming logic with the IB API. A stop-loss order, for instance, is handled by your writing code that submits the stop-loss order when your code finds that the price has moved past a particular value. What happens if you're away from your computer and Windows freezes? Or your app has a bug in it and doesn't submit the stop order? Or your Internet connection goes does? You're quite screwed! And you'll have no one to complain to about it either. I routinely use IB's API for downloading historical data. But I would never use it for anything having to do with order execution. Too much can go severely wrong!

My comments were mostly of a pedagogical nature. As I've discussed in previous posts within other topics, we often without thinking slip into forgetting the difference between percentage gains/losses and percentage "points" gains/losses. I think it's helpful to some readers to be reminded of this difference from time to time. I agree that the commissions cost as a percentage of the spread itself is an extremely important metric. I use it myself all the time. However, this metric is more directly related to risk than to profit. That is, it's a cost-of-trade parameter. The lower the commissions as a percentage of the spread value the lower one's risk. But, when closing the trade and asking how the commissions have affected your profits, it is important to focus on the commissions as a percentage of profits, not the percentage "points". For example (using unrealistic, easy calculation numbers), say you open a 1.00 straddle with commissions of 0.01. Later, you close the straddle for 1.01. Here, it would be totally misleading to state that your 1% "points" (of profits) commissions were only 1% of profits. Indeed, actually your commissions would be 100% of profits. Again, this is just something to always keep in mind when thinking in percentage terms.

Actually (using the original, non-rounded values), a 26.5% gain reduced to a 23.8% gain is a 10.2% reduction in gain. You're confusing percentage "points" with percentage itself. That is, although reducing a 26.5% gain to a 23.8% gain is only a 2.7% "points" reduction, that reduction, in percentage terms, is a 10.2% reduction. Thus, the commissions in this example cost 10.2% of the gain, which is the key percentage to keep in mind (not the mostly irrelevant percentage "points" drop).

Another way to move your funds over that might work for you that you may not have considered is to advance yourself $10,000 (or whatever you need) from a credit card, and use it to open an IB account. Many credit cards have promos where there is no fee to do this. But even if you were charged 12% annually, that would only be $100 for 30 days. That might be well worth it to you so that you can gradually close out your positions with your current account while depositing the resulting funds into your IB account. Eventually, you will have no positions open at your current account, and all your funds will be at IB. At that point (or sooner), you can repay your credit card.

My experience with IB support tickets has been bleak. Usually before I create a ticket, I've already thought carefully about the issue at hand, hoping to solve it on my own. Often I do figure out what I need to know this way The fact that I have ended up writing IB asking for help means I've given up trying to solve it on my own. So usually the kinds of questions I ask them are not run-of-the-mill. What I usually get back from IB is a reply stating that they need to ask someone in a different area and that they will get back to me. And that is where it usually remains. I very rarely get a followup response after that. If, say, a week later I write again asking if they have an answer yet, I'll get a reply saying they're still looking at the issue. If I don't check back with them in this manner, they will close out the ticket. In other words, if they sit on their behinds and don't answer, the ticket automatically closes. I guess that's one way to "solve" problems: let the ticket die. And as everyone knows, and has been pointed out in this thread, calling IB is usually not worth it at all. The people who answer the phone know much of the jargon, but that's it. They have not a clue about trading options. They know nothing about the Greeks other than how to display them in the software. Ask about any less common feature that you might wish to try and you will very likely get an on-the-fly, invented answer. I kid you not. Instead of their support simply saying they don't know the answer, they'll make up an answer for you. It's laughable! I have encountered this scenario with them many times. IB has great commissions and a very good routing algorithm (both of which keep me as an IB customer) and provides a good platform for professional traders. But you are basically on your own.

Great post. Thanks. As others have pointed out, IB's Smart algorithm has as one of its main features the ability to efficiently fill spread legs across multiple exchanges. So IB customers need not be overly concerned about this issue. The across-exchange "expanded" bid/ask price displayed is actually quite fillable on IB. By the way, there was a minor typo in the sentence following your chart: -------------------------------------------------------------------------------------------------------------- -------------------------------------------------------------------------------------------------------------- TOS would display $1.98/ $2.22 (not $1.98 / $2.18 ).

One thing to keep in mind about the IB Risk Navigator. Its calculations are often worthless. It uses for its price values the last trade value, not the bid or not the ask or not the midpoint. Why is this a problem? If the option and/or stock has low liquidity and your position has multiple legs, each leg may have traded at far different times--perhaps even different days! This makes IB Risk Navigator's resulting calculated values incorrect (actually, worthless). I've mentioned this to IB Support, and the guy I spoke with tried to convince me that this way of doing it (using the last value instead of current market value) is a better way since the tool is using real prices. When I explained again the possible time disparity (thinking he must have not heard me correctly), he didn't seem to know what I meant. So much for IB Support. Anyway, be careful when using IB Risk Navigator.

Kim, Can you please explain how you use the various volatility indexes in making your trading decisions. And why you might use one volatility index instead of another. I've noticed you refer at times to these volatility index levels when deciding whether to enter or exit a trade. What are you looking for generically with these indexes? Are you looking for absolute index values or relative percentage index values? Or perhaps momentum?

Cwerdna, I've been with IB for years, and I am very happy with their commissions and their ability to execute a trade. And these are the main features you should be looking for. Their fee schedule may appear daunting if you try to understand every sentence on their disclosure page, but it's really not. For options most of the time you'll pay $0.70 per leg. There may be "pennies" added here and there for various exchange fees, but they really are in the noise: you need not be overly concerned about them. Cancel/modify fees seem reasonable too. As far as I know, spreads never get these fees at all. Furthermore, as far as I know (someone correct me if I'm wrong), the only time IB may charge cancel/modify fees is when your (non-spread) limit order is between the bid and ask; otherwise, no fees. For example, if you have a limit order to buy below the current bid (you're waiting for the market to drop to your limit price), IB using its Smart routing system doesn't even place your order to an exchange. It just sits in IB's computers waiting for the market to move to your limit price before it sends it out so there are no modify fees then. And even when your limit order is between the bid and ask, you won't always get a cancel/modify fee: it depends on how the Smart algorithim is handling your order whether an exchange is involved or not. I rarely get any of these cancel/modify fees, and I cancel/modify often. Perhaps others have a different experience, but that's mine.

Hi Kim, I've seen you mention in various posts that you prefer straddles to strangles because the commissions are lower for straddles. I use IB as do you. Don't you just pay $0.70 per contract; thus, the commissions should be identical? I don't recall paying lower commissions on straddles compared with strangles. Or, perhaps you are referring to the fact that since an OTM strangle is always less expensive than an ATM straddle, the percentage of the price of the trade devoted to commissions is higher for a strangle since its price is lower than a straddle. Thus commissions are thereby a larger percentage of a strangle's price compared with a straddle. After writing that last paragraph, I'm pretty sure that that is probably what you've been saying. Robert

Has anyone else noticed how broken TOS's backtesting feature is? See the attached screenshot jpg file. This ThinkBack screenshot is for symbol lulu on 4/13/2012 for some of its calls. Notice how the the greeks and the implied volatility data are missing. I've noticed this with many, many different symbols while using TOS backtesting over the past few weeks and months--for options data going back over the past few years. I've written to TOS support, and they say they are fixing it. They are aware that there are tons of missing data in their database. Additionally, about a month ago I Googled this problem with TOS's ThinkBack feature and found a non-TOS trading forum containing a discussion from over a year ago concerning the exact same problem of missing data. Reading that old discussion from over a year ago doesn't give me any real confidence that TOS will actually fix this anytime soon. Worse, in my opinion, is that TOS ThinkBack software does not warn the user when an entire day of data is missing (a separate,but related, problem). When there is missing entire day data, the TOS ThinkBack software grabs the most recent day it can find even if it's a few days old, displaying it as a proxy for the later missing day as if it were valid! For example, suppose you want to know the greeks for Jan 5th for a particular option. If that day's data is entirely missing, the TOS software will grab the Jan 4th data and use it for the missing Jan 5th data. This, of course, provides worse than meaningless results. I have mentioned this problem to TOS support numerous times in the email threads I've had with them on this whole missing data subject. I've told them how really bad (I think I used the word unconscionable) it is that that they do not even warn the user when TOS ThinkBack is displaying meaningless, literally fabricated data. They have never responded to those portions of my emails. I'm a software engineer who has had years of experience with databases and handling huge amounts of data. Saving end-of-day options data into a database--although grantedly huge in amounts of data--as a database problem itself is rather trivial. I don't know why TOS has dropped the ball on this. The TOS software itself for the ThinkBack feature is solid enough (the lack of proper warnings to the user notwithstanding). It looks as if pretty good software engineering occurred during its original development. But that's of no comfort if its database is swiss cheese and provides meaningless results without warning. I'm surprised no one else here has mentioned this. Most of the options in ThinkBack do have correct price data. But for many options, as I've already mentioned, any greek and implied volatility analysis one wishes to performed is stopped cold. And even if it's not stopped cold, the fact that the displayed data could have been derived from previous days because of missing days makes much of ThinkBack's results suspect. Bottom line: TOS really needs to fix this. It would be great if more people would beat on them about this. Robert