IB's Option Model Price tick value ("ModelOptPrice") has always been buggy: it often doesn't populate properly (in TWS, or in Excel via API), and moreover can be wildly inaccurate early in the trading day. I'm looking for alternatives -- do any other brokerages or data providers offer a real-time calculated / model price for option contracts? Preferably with an API that allows them to be streamed into Excel... (A quick google search to see whether TDA / Think-or-Swim offered such a value didn't turn up anything, but can anyone confirm?)
Hi d0rian We offer a real time greeks and theoretical value feed through Tradier Brokerage. We have an API, though not an Excel API. I am not familiar with the Option Model Price. Is that something where you would feed it your vols? Our theo values are smoothed and fall between the bid ask spread > 99% of the time.
How would you calculate the option price without knowing the IV? And how would you know the IV without already knowing the bid/ask and therefore knowing the option’s price? IB and other brokers use bid/ask (mid) or last option’s price to estimate its approximate price, so really not doing much. Otherwise you could use some model, but only you’d know what model you’d like to use, so you’d have to use it yourself.
I'm a little confused by the reply, but in the case of less liquid options with wide spreads (or sometimes no displayed Bid Price or Ask Price at all), IB still calculates what I guess is more accurately described as the theoretical price (not sure if they're using Black Scholes or what; they won't divulge much info as to what inputs / model they're using.) But it's a very buggy tick value in TWS (and via their Excel API) for some reason so I'm looking for alternatives. Instead of specifying exactly what pricing model I'm looking for (which might be too narrow an ask), I figured I'd start with figuring out just what brokerage of data provider even offers an API-friendly options theoretical/model-pricing feed.
Basically I'm not sure what do you mean by "buggy" and how you yourself would compute such options pricing? Can you provide specific example/calculations of how would you calculate such option pricing? As for IB, at least in TWS they display "buggy" options pricing because I think they only use Mid and Last options prices, which aren't buggy, or are as buggy as the market provides. Basically what's "buggy" to you may not be buggy to IB or me, in the absolute terms since those prices are derived from the options pricing itself, whether that's "buggy" or not according to one person or another. And I'm not sure but suspect that their "ModelOptPrice" corresponds to what we see in IB/TWS as "Mid" or "Mark" price? Unless you see another column in TWS and which one? Or is this only available via API? At least in TWS, the "Mark" price is defined as: Any other way of computing options prices by them, could require some smoothing model (see ORATS) that utilizes surrounding options prices to determine the IV, but then it could become too arbitrary and theoretical for IB to have their customers rely on - anyone could state that those calculations are wrong, you don't like them, and you're looking for yet another model. So you're back at needing to provide your own example, before having such question resolved to your satisfaction. Any other calculations could be "buggy" according to either me or you, or someone else. Right now, if "ModelOptPrice" corresponds to the "Mark Price" in TWS then it cannot be buggy because it is clearly defined and calculates pricing according to what IB describes.
Fascinating that TWS defines "Mark Price" as it does! (I do not use IBKR, so find it curious). By that definition Mark Price for less active option contracts is useless at best! - Back to the topic of using a any option model for real time option pricing: "seems a foolish effort, as you can merely look directly at the price!" Any difference in the model price and the actual price is merely confirmation of flaw in the implementation (or usage) of the model. Perhaps you have some objective that is not clear to me!
Oh, just to be clear, I didn't use the word "buggy" in the OP to mean inaccurate or otherwise not to my liking. But rather buggy = they actually don't populate properly in TWS or via the API. (For illustrative purposes see the attached screenshot: when I activate RTD API formulas for 50 option contracts -- a pretty modest number -- all of the Ask and Close values populate properly, but the Model Prices populate very haphazardly; I usually get fewer than 10 of them to populate (only 5 of the 50 did in the screenshot), and it's a complete crapshoot as to which ones will populate and which won't; if I delete the formulas and then re-activate them, I might get a completely different subset of 5-10 Model Prices to populate, so it's not like those 45 that didn't populate have un-calculatable theoretical Model prices...they do, except the population for that particular data tick is extremely weird and inconsistent.) So yes, I understand the inherent limitations in any theoretical options pricing model, which will only ever be as good as its inputs -- whether straight Black-Scholes, or relying on bid/ask/last prices, or incorporating prices from other known Strikes/Expiries with narrower spreads or better liquidity, etc. But IB's Model prices exhibit very inconsistent and truly buggy behavior, which is why I'm looking for a different source/provider altogether.
Then maybe consider some basic ORATS subscription. I'd myself would like to be able to obtain some type of "reasonable" options pricing, but in the end I'd prefer to have more control and would/may end up doing my own smoothing after getting real-time bid/ask prices either from a broker, ORATS or IQFeed. Though the problem many people may be trying to solve is determining "reasonable" option prices at the market open, when options prices seem to be all over the place. I'd assume that's where market makers have the biggest edge since people get screwed many times they try to submit "reasonable" limit orders overnight (or submit them the day prior). While I suspect that MMs can also respond instantly to supply & demand, which could be a part of their model. Everyone else can just smooth whatever bids/asks/IV they see.
Smoothing the options surface is a very difficult task. By smoothing the IVs, call and put theoretical values can be compared to market bid-ask quotes to see if the options are under or overpriced. The smoothed IVs produce more consistent greeks by which to manage risk too. By parameterizing the smoothed curve, the shape can be evaluated comparing to other tickers or to history of the skew parameter. Here's our smooth market volatilities (SMV) process: Options quotes are cleaned, normalized and smoothed in the SMV system. The first step in the SMV System is cleaning the quotes and applying good inputs to our modified binomial pricing engine. Using ORATS popular dividend feed and option pricing methodologies, a residual yield is solved for based on the put-call parity formula. Applying the residual yield rate process helps with summarizing hard-to-borrow stocks or stocks with differing dividend assumptions. The effect is to line up the call and put implied volatilities. Next, using the call and put mid-price IVs, a non-arbitrageable smooth curve is fit through the strike implied volatilities. This smoothing system produces powerful theoretical values and accurate option Greeks. The Strikes Report shows each option's bid-asks, greeks and theoretical values. Delta, Vega, Theta, Rho, Phi and theoretical values are critical for risk management and trading. ORATS SMV also differentiates itself is by producing meaningful analytics on thinly traded securities. The SMV incorporates historical information when the current confidence in the market summarization is low. Moreover, the SMV treatment of the wings, the small delta out of the money calls and puts, produces more realistic implied volatilities than unadjusted IVs based on bid-ask prices with little premium.