Last week I wrote a little about using the Close Out button, a very useful function for forecasters who don’t want to wait for weeks, months or years for a question to settle before seeing a return. I want to expand on this a little this week, showing how you can trim your positions, taking profits or mitigating losses as you go along without totally closing out positions each time. This can be very profitable if you learn how to do it right.

First, a short recap on how our forecast markets work. With binary questions (those asking for the probability of a specific event or outcome occurring) or multiple-choice questions the scoring systems are very similar. You use your points to move the markets in one direction or the other. If the event or outcome in question occurs, the market settles at 100% and if you had pushed the probability higher, you get your points back with interest. If you had pushed it lower, you lose those points.

The Close Out button lets you cash out of a market at any point, leaving you with no further exposure. If the crowd forecast had moved in your favour up to that point, you bank a profit. If it had move against you, you book a loss but you do at least get some of your points back. The way in which the Close Out function does this is by placing a forecast on your behalf that is the exact mirror of your existing position, meaning that wherever the market goes from then on your total points won or lost would remain static.

But sometimes you might want to book a little bit of profit, or mitigate some losses without closing out of the market all together. You can do this by placing trades in the opposite direction to that which you are currently positioned.

If you do this, you will notice that the “Actual Cost” shown beneath your investment will show a negative number. In the example below, I am lowering the probability from 42% to 38% having previously been forecast that it was more likely to occur. It should cost 15 points to make this trade, and yet my “Actual Cost” is minus 15.


What is happening here is that I am reversing a part of my current position (which is more than 15 points) and so taking those points back, closing out part of my trade.

I am increasing my number of available points, shown as the green section of your points balance below, by 15. My total points, on which the leader board is calculated, is unaffected by this trade. I’m freeing up points to use elsewhere and de-risking – limiting the number of points I would win if my original forecast proves correct and mitigating the number of points I would lose if was incorrect.


I’ll give some examples next week of how to turn this into profit (some of you will be getting there already) but leave you with a great article from Nate Silver. He explains how the Republican and Democrat presidential nominees always get a polls boost from the media attention they enjoy during their respective conventions. Could you make a profit out of that information using what I’ve just mentioned? How might a strategy of selling Trump’s chances of the presidency during the Republican convention and buying them back during the Democratic convention work out?


Comments are closed.