2007 Interactive Brokers Collegiate Trading Olympiad

[Silverpike]

Lastly, I want to offer some useful info from the way last year's contest played out for me. The winner last year, Patrick Christmas, only made a handful of trades. He more than doubled his money early (2nd or 3rd week) and then basically stopped trading (or traded so few shares as to not make much of a difference in his account) probably waiting to see if anybody would catch up, which nobody ever did. This year they require at least 25 trades, which is a good rule. Also, the contest is 8 weeks instead of 10 - which in my opinion is a bad rule because I think that the longer duration of the contest, the less luck will be involved in deciding the winner.

This behavior is completely common. I have seen this in every trading contest ever offered. The problem with virtual trading contests is that it rewards users who take large risks at the very beginning. If you reason about the logic of this, it makes sense. The best way to propel oneself to the top (and stay there) is to make very large trades with lots of risk at the start.

This works very well if the contest allows multiple accounts, because you can trade in one direction in one account, and the opposite direction in another account. All you need is to choose a highly volatile instrument, like Forex or a fast moving stock (GOOG). One position is sure to win, no matter what happens. From there, you can apply the same technique again (across a total of 4 accounts), or switch to a different more conservative strategy. Many contests do allow multiple accounts, so you can see how easy it is to astroturf these things.

In the end, these contests are nothing but a red herring to me. Usually nothing can be learned by watching (or participating in) them. Because the competition uses the techniques shown above, so it really boils down to observing yourself make trades. This can be done through any regular broker, as nearly all of them will allow anyone to open a virtual trading account to practice with.

 

[Andy Nguyen]

Congratulations Gus, and Andy, well done!!

Thanks, Bridgett.

The biggest kudos go to Gus, of course. He is the man. He did this all by himself.

Before anyone congratulates me again, I need to thank Alain and point out that he is the biggest reason most of us in the competition at all.

While I'm glad I got something out of this for even trying, I have to give credits where it's due.

All credits go to Alain. He deserves all the credits for making this possible =D>
​

Thanks again, Alain. We will have a better finish next year.

 

[GusTsahas]

Thank you all!!

Thank you all for your kind words and long live the Baruch Legacy!! Congratulations to Andy and Alain and all the contestants, by participating we are all winners.

I strongly disagree with "Silverpike's" conclusion that the Olympiad contest was not worth the effort, or that it is a lottery and only a long-shot strategy will win. I entered the contest with the intent of doing as well as possible and consistently traded every day, probably placing 200 or so trades in commodities, indexes, currencies, softs, and at times, leveraged 20:1. The contest was very realistic for me as the strategy I used was based on similar ideas I have traded before and while my leverage was higher then I use, keep in mind that the top trading firms are now using 15:1 leverage.

The contest has value as the contestants gained experience with real-world trading platform, market executions and an API used by many sophisticated traders, all for free!! While we don't now if the winner was "good" or just "lucky", it really doesn't matter, each one of us got something out of this effort and we are better for it.

On to the party!!! :smt006

 

[Bridgett]

Gus,

Thank you very much for sharing your insights... I'm now thinking about entering next year's because of your example, inputs, and encouragement.

Lots appreciation!!

Bridgett

 

[woody]

While we don't now if the winner was "good" or just "lucky", it really doesn't matter

And as we've heard before, it is better to be lucky...

Good job Gus and Andy.

 

[rholowczak]

A tad late to the party...Some notes on the Wasserman Trading Floor / Subotnick Center

So we need to use their API, their demo trading system to test.
We at Baruch have a trading floor so we should take advantage of it. Somebody please ask Dan to hook us up with Prof. Richard Holowczak at the Wasserman Trading Floor/Subotnick Center.

Folks:
I admit I am late to this party - I read the postings only sporadically but will try to pay more attention in the future.

Let me make a blanket statement: Our resources at the Wasserman Trading Floor/Subotnick Center are here for all of you to use and to take advantage of. Please feel free to contact me if you think I can be of assistance in the contest or in terms of furthering your understanding of markets in general.

Regarding the technology in the Center, we have 40 workstations running Reuters 3000XTra and TradeStation, two Bloomberg Terminals and a software development classroom with MS Visual Studio (.NET 2005 edition), Sun Java and Oracle software development tools. We also have access to the Reuters Adfin API, Bloomberg API, and the Interactive Brokers API. In terms of historical data, in addition to what Bloomberg and Reuters provide, we have access to Compustat, CRSP, NYSE TAQ (both locally and via WRDS) and approx. 2 years worth of equity options data (the full OPRA data feed which is ~30 terabytes).

For a basic understanding of Equities, Futures, Options, FX and Fixed Income markets, I suggest that you attend one of our free workshops (Visit our home page http://zicklin.baruch.cuny.edu/centers/subotnick/ and click on the Sign Up For Trading Floor Workshops link.) All of the tutorials are available via our Blackboard community.

I would also be happy to hear from you if a small (or large) group of you would like to arrange a session on a specific topic or to run through a trading simulation.

Cheers,

Rich H.

 

[Andy Nguyen]

This is awesome news, Prof. Holowczak

I knew Bharath worked with you during his IB trading in 2006. We didn't quite utilize the vast amount resources at Baruch as much as we should have. Some of us attended most of the Reuters, Option, Equity workshops but didn't go further than that.

Since the 2007 IB Trading Olympiad is over, we learned a thing or two. One is that we need to understand the market very well do win.

Would you be able to provide us with some private workshop where we can create paper account and trade through the IB's API ? I'm intersted in the Forex and future market but are willing to learn about other markets as well.

Instead of using IB's paper account to backtest, can you set up so we can have our own paper account and test with the historical data you have in the Subotnick trading floor ?

Many thanks

 

[Andy Nguyen]

IB just updated the list with University name. It's nice to see Baruch College 2 years in a row.
http://www.interactivebrokers.com/en/general/education/IBTradingOlympiad.php

 

[Chris]

Hey guys,

I'm not sure if this was posted before but here is an article from the Financial Times on this. It talks a little about the trading strategies used from the winner and Gus.

http://www.ft.com/cms/s/13ce9470-dbe2-11db-9233-000b5df10621.html

Thanks,

Chris

 

[HEMT]

congratulations

 

[Maserati]

Ok so it says "Trades may be based on any combination of technical or fundamental data" but can someone please tell me exactly what data is given? I'm trying to prepare for the next one...

Thanks

 

[alain]

You can retrieve any data available through TWS.

 

[Stefan Zota]

It would be very interesting to get a feel of different algorithms.
More specifically, as it was already mentioned, since risk is not part of the equation, a lot of instruments will be neglected. So, a lot of theoretical side would not be used to create the model.

It would be also interesting to know the complexity of the code. Andy was mentioning 1 month of programming. If this means, at least 3-4 hours a day, then the final size is reasonable (7-10k lines of code) ...

 

[alain]

It might be even simpler... but, word of caution, there might be risk levels in next year's competition. I spoke with Mr. Wilkinson and they are looking to put some VaR limits into the game.

 

[GusTsahas]

I've said this before, and at the risk of being repetitive I'll say it again, everyone's time and programming skills will be most productively used in testing various trading strategies first and coding to the IB TWS interface only during the last couple of weeks.

Leave the interface code for last... I understand people are more comfortable programming to a problem with known parameters like communicating with the IB API, but you cannot make money "communicating", only by trading!! Concentrate your effort into investigating trading strategies that work in different market environments and backtest, backtest, and backtest some more..... Your strategies need to make money in paper simulations, as the actual trading contest is harder with issues having to do with unforeseen mistakes, problems with fills, bad days... but at least you started off with a "positive expectation" strategy.

Have a good semester everyone, GUS

 

[DominiConnor]

Just to underline Andy's point...
Although $25K has some utility, it is not the prize.

The prize is being able to say to a top tier employer "I beat a lot of other people in a trading competition".
That's worth far more than $25K. If I were forced to price that, would be more than everyone you are related to by blood earned this century.