Monday, June 24, 2013

Why Arbitrage Fails

We're back, after a brief break. I'm going to write this post as promised, but then take a slightly different turn with this blog. I'll explain what's happening and my reasoning in the next post.

So, let's talk arbitrage. There are several sources of arbitrage in the MTGO secondary market. Human-to-human, human-to-bot, bot-to-human, and bot-to-bot. To clairfy, human-to-bot is where you buy from a human and sell to a bot, and make the difference. I can tell you now that none of these are worth the time. It can take quite a while to find a buyer and a seller, of which the buy price is higher than the sell price, and where both parties are guaranteed to buy or sell, respectively.

This is the first risk. Since you don't know all the information on either side, you take on appreciable risk when you attempt to conduct arbitrage. Your buyer might decide not to buy after all, or your seller might decide to keep their cards. The latter is the preferred scenario, since then all you need to do is apologize to the buyer and move on. The former is the more problematic; when you can't actually sell a card, it becomes deadweight to you. You've lost tix buying it, and if you sell it to a bot, it will be drastically undervalued. Bots always undervalue what you sell to them and overvalue what you buy; it makes basic economic sense to them.

Now, this risk could be worth taking on, if it were met with equivalent reward. But in the market, the reward is extremely limited by the game itself. Players generally find more than four of a card (a playset) to be useless, and bots will generally not hold more than four or eight copies of a card. This means that, unlike stocks, profit is very limited. Even if you find a few bots to sell to, you're not moving more than twelve copies of a card for more than 0.25 tix of profit each, and even that's a scenario I've never reached. A more reasonable scenario would be moving two copies of a card to one bot at 0.1 tix of profit each, after 10-20 minutes of searching. Definitely not worth the time.

In addition, remember, that 1 tix is the smallest amount of currency you can move from bot to bot. Though bots hold credit, buying two cards at 0.3 tix each and unloading them to two different bots for 0.6 tix each still loses you one ticket since the original bot will require one ticket to buy the cards in the first place and hold 0.4 credit. Over many transactions you might end up with your net tix in the black, but in the short-term, it's not worth it. It's not even worth it in the long-term really, when you weigh it against how much time you spend on it.

A final source of risk comes from convertibility. Credit is all well and good to have on bots, but ultimately you want to be able to convert those credits to tix. This further limits your field as you have to ensure every bot chain your sell to has tix with which you can cash out, or comparable value cards. Sometimes, if you sell at a grossly overpriced rate to a chain (e.g. runo7ke, I think the bot was) all their cards will also be grossly mispriced, and on top of that, have no actual value.

One thing I have yet to try is card speculation. What I've done so far is more akin to day trading, but I imagine buying cards and holding on to them for months could be profitable in the long-term. You could use a site like (I actually don't know of another like it!) to watch your prices and follow trends. I'm thinking about doing this in the future, but right now, I want to wait until the release of M14 and maybe even Theros to see how booster pack and card prices are impacted by both new set releases and cards phasing out of standard before I enter the market there.

But in short, making tix through arbitrage isn't worth it because even without considering risk, the time consumed vastly outweighs the reward.

No comments:

Post a Comment