your interview on 'Chat with Traders' has introduced me to you and I have been fascinated by your character and career ever since. Alone the interview + this extra write up shows that you care about the community.
I (just a student) am very interested in doing some analysis regarding the options bids and offers in real time or historically. I would like to know which (German) broker API you used was the simplest and most user-friendly back then? I would be very grateful for a short answer as it would help me a lot.
One more question: Why would you have needed to use several brokers for the retrieving of warrant prices? For trading it's obvious: take advantage of zero-cost trading months, differences in trading fees, difference in execution rates ect.
But shouldn't the quoted price for a particular warrant be the same across all brokers or did/do the dealers (in this case the banks) quote (slightly) different prices to each broker?
Hi, I saw the interview with Aaron. its one of the most practical, in depth and informative videos I have ever seen on YouTube. I went through your write-up and request to provide your comments on below questions:
1.What made you certain that model assumptions in DAX Warrants and Eurex Options were same and you weren't missing out/miscalculating some implied assumptions dealers might have put in DAX Warrants? was it keeping record of relative mispricing and observing it converging after few seconds to minutes made you sure that modelling is same and it is execution inefficiency and you are not missing out any other variable?
2. I checked the statistics you gave above but somehow not able to interpret it well. Can you please help me explain what was expectancy or edge of system?
3.In cases where you had losing trades(except the outliers happened due to being trapped in dance), what were the reasons for it? was it due to underlying market/volatility moving against you?
1. The payout of a warrant is clearly defined and described in the offering documents. The final settlement price of a warrant was determined slightly differently than for an ODAX option, but this is a tiny detail which doesn't matter for pricing or modelling. The warrants that expired on the same day as the ODAX options, 3rd Friday of the month, had the same financial payout and therefore needed to have the same price except for credit risk. There are also warrants that expire before or after the 3rd Friday and they would sit on a different point on the term structure. None of this mattered for the process of determining relative cheapness/expensiveness during the day.
2. Expectancy, or edge, tells you the PL expectation for every trade. Calculate the PL of your average winning trade and the PL of the average losing trade. Then figure out the % of winning vs. losing trades, and then you can calculate the PL expectancy per trade.
3. Yes. I entered at what I thought was a low/attractive price and then the market turned around to the upside and vols came down. As a result, the dealers were under no pressure to remark their vols higher, or only a bit, in which case I quit for a small loss.
Thanks for the reply. for point 2: I understand what edge is and how to interpret it, but I am unable to interpret it from the statistics format you have given. please correct me if I am wrong, does Avg. Spread mean average winning %(for example avg win of 15% equates to avg spread of 1.15)? shouldn't we also have avg losing spread to calculate edge of system? I am not able to figure out the expectancy of the statistics you posted above, request you to please help me regarding it.
When the B/O spread on a warrant is 1 cent, the total spread cost per side for 10K warrants is 10K x 0.01 EUR = 100 EUR. Thus, for a strangle consisting of 10K calls and 10K puts the cost is 200 EUR, again assuming a B/O spread of 1 cent for each. Let’s now assume that the PL of a trade is +800 EUR. In this case the PL is 4 spreads (800 EUR / 200 EUR). In other words, I made 4 B/O spreads on that particular trade. That's a way to normalize the PL.
Hello Moritz (Seibert),
your interview on 'Chat with Traders' has introduced me to you and I have been fascinated by your character and career ever since. Alone the interview + this extra write up shows that you care about the community.
I (just a student) am very interested in doing some analysis regarding the options bids and offers in real time or historically. I would like to know which (German) broker API you used was the simplest and most user-friendly back then? I would be very grateful for a short answer as it would help me a lot.
Best Regards and wishing you a nice Sunday!
I used the interfaces of several brokers to retrieve warrant prices: Vitrade, Flatex, Onvista Bank, and some others. I used IB for ODAX and FDAX refs.
Thank you for the answer Moritz!
One more question: Why would you have needed to use several brokers for the retrieving of warrant prices? For trading it's obvious: take advantage of zero-cost trading months, differences in trading fees, difference in execution rates ect.
But shouldn't the quoted price for a particular warrant be the same across all brokers or did/do the dealers (in this case the banks) quote (slightly) different prices to each broker?
Their interfaces all had different latencies... this was important.
Makes sense. Thanks a lot!
Hi, I saw the interview with Aaron. its one of the most practical, in depth and informative videos I have ever seen on YouTube. I went through your write-up and request to provide your comments on below questions:
1.What made you certain that model assumptions in DAX Warrants and Eurex Options were same and you weren't missing out/miscalculating some implied assumptions dealers might have put in DAX Warrants? was it keeping record of relative mispricing and observing it converging after few seconds to minutes made you sure that modelling is same and it is execution inefficiency and you are not missing out any other variable?
2. I checked the statistics you gave above but somehow not able to interpret it well. Can you please help me explain what was expectancy or edge of system?
3.In cases where you had losing trades(except the outliers happened due to being trapped in dance), what were the reasons for it? was it due to underlying market/volatility moving against you?
Thank you for the kind words.
1. The payout of a warrant is clearly defined and described in the offering documents. The final settlement price of a warrant was determined slightly differently than for an ODAX option, but this is a tiny detail which doesn't matter for pricing or modelling. The warrants that expired on the same day as the ODAX options, 3rd Friday of the month, had the same financial payout and therefore needed to have the same price except for credit risk. There are also warrants that expire before or after the 3rd Friday and they would sit on a different point on the term structure. None of this mattered for the process of determining relative cheapness/expensiveness during the day.
2. Expectancy, or edge, tells you the PL expectation for every trade. Calculate the PL of your average winning trade and the PL of the average losing trade. Then figure out the % of winning vs. losing trades, and then you can calculate the PL expectancy per trade.
3. Yes. I entered at what I thought was a low/attractive price and then the market turned around to the upside and vols came down. As a result, the dealers were under no pressure to remark their vols higher, or only a bit, in which case I quit for a small loss.
Thanks for the reply. for point 2: I understand what edge is and how to interpret it, but I am unable to interpret it from the statistics format you have given. please correct me if I am wrong, does Avg. Spread mean average winning %(for example avg win of 15% equates to avg spread of 1.15)? shouldn't we also have avg losing spread to calculate edge of system? I am not able to figure out the expectancy of the statistics you posted above, request you to please help me regarding it.
When the B/O spread on a warrant is 1 cent, the total spread cost per side for 10K warrants is 10K x 0.01 EUR = 100 EUR. Thus, for a strangle consisting of 10K calls and 10K puts the cost is 200 EUR, again assuming a B/O spread of 1 cent for each. Let’s now assume that the PL of a trade is +800 EUR. In this case the PL is 4 spreads (800 EUR / 200 EUR). In other words, I made 4 B/O spreads on that particular trade. That's a way to normalize the PL.
that gives us winning multiple, what about losing one? shouldn't we consider avg losing spread multiple to calculate edge?
The 1.26 spread I show in the stats includes winning as well as losing trades.
oh okay, I got it. Thanks!
I used the interfaces of several brokers to retrieve warrant prices: Vitrade, Flatex, Onvista Bank, and some others. I used IB for ODAX and FDAX refs.