Different data, different results

For over 50 years I have been doing mathematical analysis on a variety of subjects including sports statistics, stock investments, large mainframe computer performance, and more lately cryptocurrencies. I’ve learned that mostly the results are surprising and don’t support your thesis going in. That is the case in my latest analysis.

In my previous post I reviewed an analysis I did last summer on two simple cryptocurrency trading strategies. One was a buy and hold. The other was to sell a portion of the currency with the best performance for the day and use the funds to buy the one with the worst. Effectively somewhat of a re-balancing daily. There were 3 currencies involved BTC, ETH, and LTC. The buy and hold outperformed the re-balancing approach. That was disappointing but not a major surprise.

I’ve now rerun the analysis with more and largely different data. I’m using the last 4 quarters, from April 1^st, 2017 to March 31^st, 2018. There are 4 months in both sets of data. I started with $1,000 split equally between the coins and iterated daily computing the new balance. The buy and hold was simply a price change. The rebalancing required computing the changes and doing a simulated sell and buy, then computing the new balance. The resulting balance for the buy and hold was $9,532, while the re-balancing ended with $10,423. In this case the re-balancing outperformed the buy and hold.

The two tracked more closely than with the previous study. The next step is to dive a little deeper in the results to see if anything else can be learned. Follow me on Twitter, @ole44bill, to know when the next post is made.

Crypto: First attempt at a simple trading strategy

I am starting to analyze crypto-currency data again to outline future investing options. In my first effort, I tried to convince myself that focusing only on bitcoin (BTC) was better than looking at some of the alternate coins. As documented in my previous posts, I failed and demonstrated that there are some coins with better growth patterns over the last year.

I’ve decided to re-visit some of my prior analysis and update the data considering the crazy growth and decline patters in the latter part of 2017 and the first part of 2018. In July 2017, I looked at two simple strategies for trading coins. The coins involved were BTC, ETH (Ethereum) and LTC Litecoin). I downloaded their daily price data from Jan 1^st, 2017 through July 29th, 2017.

The 1^st scenario was buy and hold. Starting with $1,000 on day 1, I used one-third to purchase as many of each coin as possible. This resulted in 0.33 BTC, 40.95 ETH, and 75.41 LTC. These were held until the end of the period and their daily value grew to $12,047 at the end of the period.

The 2^nd scenario was a simple trading strategy. At the close of each day, the change in value of each coin was computed. A portion of the best performing coin was sold and the proceeds used to buy the worst performing coin. Kind of a dollar cost averaging scheme. The amount of coin depended on the amount the best performing coin changed. At the end of the period, the value grew to $8,988. This wasn’t as good a strategy as buy and hold (thank you Warren Buffet).

Following is a chart of their daily value.

The next step is to repeat the analysis using data through the end of the 1^st quarter of 2018. That will be my next post. Follow me on Twitter, @ole44bill, to be notified when I post it.

Sports: Baseball simulator efforts

This is my first post on sports wagering. It’s a bit out of sequence because I had planned on laying some groundwork for these posts with some background information. But there is an open discussion on a forum at https://www.sportsbookreview.com/. It’s in a sub-forum called “Handicapper Think Tank” (a good place to discuss handicapping ideas). One user opened a thread on “using video games to handicap sports”. This is a topic near and dear to my heart and one I’ve worked on for many years.

I’ve decided to discuss my history trying to simulate MLB games to generate the final score of many simulated MLB games that can be compared to today’s money lines. If the results are truly representative of the likely outcomes of the games, then they can be used to suggest potentially profitable wagers.

First, you must have a simulation engine that generates realistic results. I suspect video games put more effort into graphics than statistical accuracy. I tried 3 different statistical simulators. I replayed an entire season 500 times to generate 500 results for each game. I used the actual lineups and the final player stats for the year.

I then used the results against the betting lines for each game generating pseudo bets for those with a positive expected value based on the simulations. I accomplished two things by doing this. One was to verify that the simulation engine produced reasonable results. And the other was to see if this could be profitable. The answer to both was yes. The selection criteria for picking the simulator to use was more involved than simply picking the one with the best wagering results. Just as important were things like being able to enter and use player stats and speed of generating the results.

But these results were optimal because they used the actual season stats and lineups. To simulate a new season's games, you must decide what stats to use for the players and what lineups to use. To decide what stats to use I embarked on a major effort of generating player stats to the day of the games for the previous seasons. Thus, to simulate a game on May 2nd, I would only consider stats for players through May 1st. I automated this process as much as possible. It required running my PC's 24 hours a day for weeks. I tried using YTD stats, the last 2 weeks, and the last 3 years. I learned using more data was better than using less.

This then led to a project to build player stats files to forecast a player’s performance and as the year progressed, blended these with YTD stats. I used randomly selected lineups from recent games in the morning and simulated each game 5,000 times. When the actual lineups became available later in the day, I used these and ran another 5,000 simulations just before game time.

I used this simulator for several years and made real wagers on the results. As expected the results were less than the optimal ones using previous seasons tests. I did not find this to be profitable and so after several years abandoned the effort.

I’m not convinced that this type of approach is doomed to failure. Simply put, it’s a lot more work than one would expect, and I couldn’t get it to work. I actually dream about revisiting this project again in the future, but it’s unlikely in the time I have left.

I hope to get back to a more orderly sequence in my next sports blog post. Follow me on Twitter, @ole44bill, to know when these posts occur.