Mods to Competitions

I have made some minor additions to the Trading Competitions feature.



In response to your suggestions, I have changed the Web site to allow people to join competitions after they have begun (or not - the setting is customizable by the Competition Organizer).



Also, organizers can specify a minimum number of required trades per competition. This solves the problem of lucky gamblers betting their entire wad on one trade and hoping for the best.



Now all that needs to happen is that someone has to start a competition. Then we can see if this pig flies.



Matthew

I don’t think setting a minimum number of trades per competition will solve anything. All a person has to do is buy something then immediately sell it and they’ve completed a trade. Just do that as many times as needed to meet the minimum. At $40 a round trip, doing it 10 times is $400 which is pocket change for a $100,000 account. You could set a max % of account size per trade, to try to beat the lucky gambler problem, but they could just do multiple trades of the same security. As someone said before, most any restriction can be gotten around.



I have some questions about the trading systems that have to be created for the competitions.



1) Can people subscribe to them?

2) What happens to the system when the competition is over? Do they convert to regular trading systems that can continue? If so, what is the pricing structure?

3) Do the systems created for the competition affect the owners collective2 rating?

Tim is right, one can make 1 huge trade and a bunch of small ones and the min. requirement is fullfilled. But the % requirement works better if only with stocks.

With options, even the % max. doesn’t work because if let’s say 20% is the max. that I can commit to one trade, I just trade similar options, that usually move together (like QQQ, NDX,SOX).

If the % max. is signifficantly low (like 10%), that gives a headache for the lucky guesser.

But this feature still should be added, because it prevents stocktraders from having one lucky guess.

Tim is wrong on the multiple trades of the same security because the trader than have to be right multiple times. Obviously the program wouldn’t let to buy the same security if the % limit is already reached.

So in short, % limit is good, number limit is ineffective.

Let me answer some of your questions, but then ask a few of my own.



First, what we’re trying to avoid is “Lucky Bastard Syndrome” - the possibility that a competitor opens one position, hopes for a huge win, and then wins the competition. We are not trying to prevent “Conservative Cash-Is-King” investors from winning. If a person opens and closes 10 trades very quickly, this is the same, essentially, as keeping all your money in cash. Might you win with this strategy? It is possible, but if the Trading Competition is filled with good investors, probably not. By setting a minimum threshold, we - in theory, anyway - prevent people from letting everything ride on one big trade. I know, I know - you can game this, too, by meeting your trade requirement with garbage trades, and THEN letting everything ride on one trade. But before we throw in the towel and decide fair competition is not possible, let’s see how it works out. I suspect that people are not out to game the competitions - the people on this site really want to win fair-and-square. If we find a consistent problem, there are ways around the issue - perhaps a vote at the end of the competition to see if everyone agrees the winner played fair? Or maybe just a “Matthew’s Word Is Law” judgment from on-high. In any case, I think we are worrying a lot about a problem that may not be much of a problem, after all.



Now, about your questions, Tim:



First, systems that have been entered into a competition ARE NOT part of the C2 score calculation. For calculation purposes, they are ignored.



After the competition, your competition sytems become inactive - they will be visible, but you will be unable to enter any new trades. As of now, you can’t roll these systems into your other long-term systems you manage on Collective2.



Finally, regarding subscriptions: right now it is possible for people to subscribe and pay for subs to competition systems. This is not really a conscious design choice; it is simply a result of the way the C2 software was written. But I have an idea/question about this.



What do people think of the following idea: that we allow anyone to subscribe to competition systems - for free - so that everyone can see the open trades? The alternative idea (which I don’t like as much) is that we let only those people within the competition subscribe for free to competition systems.



Does anyone have any thoughts about this? Any reason why we wouldn’t want this to happen? It might make competitions more exciting and prompt more people to join…



Matthew

I meant multiple simultaneous trades, such as with a 10% limit per trade:



trade 1) Buy 1000 shares xyz at $10

trade 2) Buy 1000 shares xyz at $10

trade 3) Buy 1000 shares xyz at $10

trade 4) Buy 1000 shares xyz at $10

etc.



It would have to be something like no security and all its derivitives may equal more than X% of portfolio value at purchase time. One problem I see with doing it this way is that it would unfairly work against people who want to do strategies such as spreads and covered calls. Since Collective2 doesn’t support spread orders or buy-writes, a trader would have to leg in to the trades and doing so would severly reduce the number of shares/contracts they should be allowed to procure according to the % rule because they are legging in to offsetting positions. How would credit spreads and selling naked puts be treated as far as the % of portfolio? Use the margin amount to determine % of portfolio?

I do like your idea of making competition trades visable to everybody. At the very least it has advertising advantages, and getting more people to this site has to be one of your major objectives.

The whole point of the buy/sell, buy/sell immediately example was to show how easy it is for anyone to get around the supposed fix for the “Lucky Bastard Syndrome”. I don’t like whole minimum # of trades idea in the first place. What if a trader wants to put 20% of account value into each of 5 stocks, then sit on them for the whole competition, but the minimum number of trades is 10? I don’t think he should be penalized because of his style. He paid to play. I vote to remove the min # trades setting.



I don’t think anyone should be able to subscribe to the competition systems for free. It’s a competition. That’s like playing 5 card draw with your cards showing. That said, I think each owner of a system should have the choice as to whether they offer their system to subscribers or not and what the fee is. The reason being is what’s the point of paying for regular trading systems when you can get them for free in the competitions. Just look for a good consistent trader and subscribe to his competition systems for free.



I think all subscribers of a particular regular trading system should be able to subscribe to the competition system offered by the system vendor for free after the initial trail period for the regular trading system is over, as long as they are still subscribed to the regular trading system offered by the same system vendor after the initial trial period, if any.



I think the minimum number of trades is an important variable parameter to judge any system. Any statistics is not considered valid, unless it has atleast 30 trade samples in the time period on which the system is tested. This is the way the Expectancy of a system is calculated. If a system does not generate enough trades, it is not possible to accurately judge a system from statistics point of view.



That being said, the ideal number of trades for a competition could be 10 trades/month for the monthly results and 30 trades/quarter for the overall (quarterly) results to judge the overall winner assuming that the competition is conducted monthly for 3 months and the overall winner is judged at the end of the 3 month period. This way even if a good trader does not make 10 trades/month, he still has a chance to be the overall winner if he makes 30 trades/quarter and emerge the overall winner. This facilitates both the day traders and longer-term traders and gives both chance to win the overall competition which should be the goal. We could also have 30 trades/6 month period or 30 trades/year or 30 trades/decade (and judge the overall winner accordingly for each period) if the competition continues for that long which would facilitate the likes of Mr. Warren Buffet.



This also eliminates the lucky winner who may win that particular month and should be allowed to do so, but may not emerge as the overall winner, because he has to make 30 trades/quarter and no scalping (entering and exiting immediately just to meet the minimum # of trades requirement) is allowed.



rgds, Pal