Exchanges Part 2

 I was thinking about my last post and realized that I had a few more things to say about using an exchange, market-making in general and a little bit more on NoVig specifically.

This will get into the weeds a bit. If you do use an exchange or are any kind of arber at all, you'll get something out of this. If not, you might want to wait until my next post.

I have made one subtle but very important change in the way I set up big arbs/scalps/half scalps. This is maybe obvious to everyone else but I just learn on my own as I go so I don't really know what the average arber does or how he goes about things.

The way arbing typically works is first you need to find something arb-able. Let's say I can bet a prop Over at +110 at one place and the Under at -105 at another place. One of these lines has to be fairly off market. Let's say the +110 is off market. I would max bet the +110, let's say the max bet is $1k. If your bankroll justifies it and you are confident that the +110 is indeed off market, your best play is to just let that $1k stand. However, with props especially, you can't always trust the market. So even if your bankroll does justify being naked with a $1k bet, it can oftentimes make sense to buy some back on the other side at -105. This will make your bet have better effective odds. Say you bet $1k to win $1.1k on the over at +110 and $525 to win $500 on the under at -105. Now your effective bet is Over $500 to win $575, or +115. The more you buy back, the better your effective odds but the less volume you get down.

This is pretty standard, straight forward stuff so far and something we've covered here before.

However, here is where I made a change.

Let's stay with the example above. Let's say I made both those bets and now have the 500/575 (+115) bet on the Over. In the past, I might arb that bet with something at, say, -112. After all, -112 is arb-able with +115. But when you do that, all you're really doing is cancelling out the bet you made at +110. 

Now, when I'm market making on NoVig, I will never take any bet that isn't better than my worst bet on the other side.

For example, say I get -104 matched on an MLB Team A and +110 matched on their opponent, Team B. The worst possible number I would take on Team B now would be +106, and the worst number I would take on Team A would be -108. No matter what. I never take any bets that are not a pure arb with the worst number I got on the other side. 

This is probably common knowledge and I did generally operate like this in the past, for the most part. But now I'm way more strict about it.

I think I learned this lesson from trading stocks/crypto. If you buy a stock at $10/share and then sell some at $20/share, you really are just locking in a loss if you buy back in at $21 or higher. I know there are other considerations with stocks like taxes and liquidity needs and whatnot, and sometimes things change and buying back in higher is the right play. But then that means that by definition, your sell at $20 was a mistake. It cannot possibly be true that selling at $20/share and buying at $21 or higher are both correct. One of those moves has to be a mistake.

Here's a good rule of thumb I think of often. If you could repeat whatever you are doing an infinite amount of times, would you make or lose money? If you buy a stock at $21 at sell at $20 over and over, you will lose all of your money eventually. If you do the opposite, buy at $20 and sell at $21, over and over, you will eventually have all the money in the world.

Same thing applies to arbing. If you can bet one side at -104 and the other at +106, over and over again, you would eventually win all the money. Literally.

So now I only take a bet if it is arb-able with the worst number I got on the other side. This also has the side-effect of making me way more picky about what I will even offer in the first place. 

I have mostly operated like this forever, but seeing it in a different form with stocks really crystalized the idea for me, and same thing with NoVig. Seeing all the bets in each arb very cleanly has nailed the point home for me. I used to only care about my effective odds. To me, a half scalp where I have +115 effective odds was no different than a regular bet at +115, which is still true. But now I realize that that isn't the whole story. You have to consider the bets that got you to +115 in the first place.

A couple tips about NoVig specifically that I forgot to mention:

One of their 'challenges' is to log in every day for a month straight. All that means is clicking on the app so it's pretty easy. However, you have to actually go into the challenge and click on 'claim points' at 5 days, 15 days and I think 25 days to actually claim it. Honestly it's barely worth it but if you're on there everyday anyway, you might has well know that.

Also, it can be a little tricky to use at first. You're offering bets on the opposite side of what you want, so it is possible to make mistakes. One time I put a plus sign on my offer when I meant to do a minus sign. So I got some bet at like -150 when it should have been +150, which is obviously a disaster. Honestly, I'm surprised that doesn't happen more often. NoVig does a neat little thing where they show you the best price you can offer and that's actually a good tool and a good way to not make a mistake like that. I am always sure to look at the market after I make an offer to make sure nothing looks weird. 

Lastly, always remember that this shit is zero-sum. You can only win if someone else loses. Winning money on an exchange long term is hard. Literally everyone else is trying to do the same thing (well I suppose that isn't exactly true. Some people do use them to lay-off action and as the hard sides of arbs so I guess there are people trying to lose on there. But they're winning that money and then some back in other places. So even they are taking money out of the whole ecosystem.) My point is that it isn't easy. Having any kind of profit long-term is commendable. Even the way I do it, offering markets on both sides of a game, is nowhere near fool proof. Plenty of times I'll get only one side filled and will be naked on a bet. Sometimes it'll be a really bad number too, if the market moved hard one way and I was too slow to react. So if I can maintain a 2% ROI on there, I would be pretty happy. I think 2-3% is about right. It'll be interesting to see what my ROI is after the NFL season. 

That's all for today, folks! Check back soon, bye for now!