Posted by Ryan

Recent events have helped to reinforce the most important aspect of sports betting: The number you bet means everything. Most novices are looking “for the winner”, but anyone with a grasp of probability knows that anything is possible. Therefore, the odds you bet are all that matter.

As such, I’ve been trying to come up with a bet posting strategy that doesn’t involve people searching for numbers they will never find. Not only does this waste your time when you’re trying to find a good bet, but it wastes my time as no one is getting any good out of it.

So here is my plan: I will post any bets I make that I expect to be reasonably available. Also, I’ll only post bets that are worth a decent amount. The first few days of MLB have shown that most bets are for very small amounts, thus it seems pointless to me to post bets for small amounts that maybe one person will be able to find.

I hope this explains why I haven’t posted many bets over the past couple of days.

Just remember: if you can’t find the same or better price than the one I’ve bet then you are probably better off passing. In fact, I would expect this to be the norm for most of my readers. I know some of you have plenty of outs to work with, but if this does not describe you then please don’t get frustrated if you can’t find a good number.

If you're new here, you may want to subscribe to my RSS feed.
Thanks for visiting!

Posted by Ryan

After last night’s losses I’ve now dropped almost 25% of my bankroll so far in the month of March (23.5% to be exact). In other words, I now have about 75% of what my starting bankroll was on March 1st.

I hate to lose, so this made me wonder: just how likely is this to happen?

Based on my simulations, and using the Tampa Bay game from last night as a guide, I am going to drop 25% of my bankroll about 21% of the time when betting with a 1/3 Kelly stake. This drops to roughly 17% and 12% for 30% Kelly and 25% Kelly, respectively.

Therefore, I shouldn’t be all that surprised that I’ve dropped 25%.

Hopefully you aren’t all that surprised now either.

Posted by Ryan

Making multiple bets while keeping within the Kelly criterion is a tricky proposition.

If you have no idea how to use the Excel optimizer (like me) then consider using Ganchrow’s calculator to calculate the proper Kelly stakes for multiple bets. Although I’ve created a Kelly criterion calculator, it is very basic and does not support multiple bets.

Using Ganchrow’s calculator is fairly simple:

1. Select the number of bets you’re making
2. Enter the Kelly Multiplier to use (1 for full Kelly, 0.5 for half Kelly, 0.25 for quarter Kelly, etc.)
3. Enter the dollar amount of your bankroll next to Starting Bankroll
4. Enter the odds and probabilities associated with all of your bets in the boxes provided
5. Click the Calculate Kelly button

Once you follow these steps, the amount to risk on each bet will be listed in the box provided.

To see the percentage of your bankroll to risk, simply change Starting Bankroll to 100% and click the Calculate Kelly button.

See, that isn’t so hard, is it? I hope to create my own calculator one day that allows for greater flexibility, but until then Ganchrow’s calculator is a great way to approximate Kelly stakes for multiple bets.

Posted by Ryan

In case you haven’t been paying attention, I’ve won 22 of my last 23 bets. This is by far the best streak I’ve ever had betting sports, but this is a very rare occurrence.

A streak like this will unfortunately feed the notion that betting sports is easy and that there exists some magical method to “finding winners”. This was confirmed by an e-mail I received from a reader this weekend. In short, he wants to know what “the secret of [my] success on NHL” is.

The reality is that the magnitude of success I’ve had can only be attributed to luck. While I expect to profit, I do not expect to win 22 out of 23 bets. Every bet I make has a probability of losing. I’ve never made a bet that has a 100% chance of winning, and I don’t expect to find any such bets in the future.

What I do know is that I can expect to have losing streaks that will feed the notion that I’m one of the worst sports bettors out there. When that eventually happens I’ll reference you back to this post.

Keep your expectations realistic

You should always keep your expectations realistic, especially if you’re new to betting sports. As much fun as it is to get lucky, it is not what you should expect. Always remember that hard work wins out over the long term.

Posted by Ryan

If you’re not familiar with the Kelly criterion then I suggest you read my money management guide before proceeding.

Up until recently I used a 50% Kelly fraction when sizing my bets. I did this because I thought that 50% of full Kelly was good enough to reduce my volatility yet still make a good profit.

That changed, however, once I got a better grasp of the risk associated with betting various fractions of the Kelly criterion. It all started when I read the Kelly FAQ at BJMath.com.

The FAQ suggests you look at the probability of doubling your bankroll before losing half of it as a guide to choosing the Kelly fraction that fits you best. Seeing the numbers (the table from Q9) encouraged me to run simulations in regards to sports betting to see how these probabilities fit my expectations.

The table below shows the results:

Note: the Probability column denotes the probability of halving ones bankroll before doubling it when using the listed Kelly Fraction.

The simulations used to calculate the data in the table above assume an event with a 55% chance of winning against odds of -110.

As you can see from the table above, a larger Kelly fraction greatly increases the risk to halving your bankroll before doubling it.

The world isn’t perfect, and it takes a really long time to have full Kelly be superior to other options. As such, I’ve chosen to scale back to a roughly 33% betting fraction to fit my risk profile better.

You should be using the Kelly criterion

If you’re not doing so already, you should be using the Kelly criterion to size your wagers. Read my money management guide and pick the appropriate Kelly fraction from the table above and start profiting.

Posted by Ryan

Let me first say that I highly suggest you use a Kelly criterion-based money management plan to optimally manage your money for sports betting.

That said, the intent of this post is to point out how making any real money with a Martingale betting strategy is impossible.

The Martingale (betting system) page at Wikipedia does a good job of showing the math behind a martingale betting strategy, but I’d like to expand on that with some actual data to show why this strategy just doesn’t add up.

Let’s say you want to bet NBA point spreads for profit without having an edge, thus you simply choose to flip coins as a way to choose a side.

This means your probability of winning is 50% against odds of -110. To make a profit you will need to start by risking 1.1 units and doubling until you win.

How much you stand to lose

Showing how much you stand to lose really hammers home why a martingale betting strategy is a terrible idea. For this discussion I am disregarding betting limits, as that makes even small losing streaks tough to withstand.

The table below shows how small losing streaks can really burn through money as you try to double up to make a profit:

This table assumes you will continue to bet a multiple of 1.1 units after each loss. If you risk more so that all bets are to net 1 unit of profit then your actual loss will be higher.

While these losing streaks aren’t likely, the more you bet the more likely you are to have one of these streaks.

A better investment

If you’re thinking about using a martingale betting strategy, I hope I’ve encouraged you to invest your money in a better way. The first such investment should be in developing ways of finding profitable bets.

Once you do this you can use a smart money management plan that will help you optimally bet your money, as there are no bet sizing tricks that will guarantee you a profit without an edge.