winning percentage Archives - Algorithmic and Mechanical Forex Strategies

The guy that bet on Leicester City every year

September 5, 2016 by Shaun Overton Leave a Comment

Leicester City started the 2015 season with terrible odds of winning the Premier League Championship. Bookmakers only game them odds of 5,000:1 of winning.

To put that in context, you are more likely to die riding a bicycle than you were to win a bet on Leicester City. Or, you can think of betting on Leicester City every year. If you bet on them every single year for 5,000 years, you would expect them to win a grand total of… once.

2014 was hardly an indicator of their pending success. They were nearly relegated to a lower division (i.e., kicked out of the Premier League). And yet, they did win the championship last year.

Leicester City’s Biggest Fan

Meet John Michklethwait. He’s the former editor-in-cheif at The Economist and he’s currently editor-in-chief for Bloomberg. Clearly, he’s a very smart man. And yet, despite the odds and repeated disappointments, John bet on his old love, Leicester City, every single year dating back to the 1980s. That’s roughly 30 years of nonstop losing.

It wasn’t a lot of money each year: just £20. We all have our indulgences. I see the value of having skin in the game. £20 on a season is enough to make one care, but not so much that he’s upset about losing it.

Then something disruptive happened. John moved to the US last year for his position at Bloomberg. The chaos of the move threw him out of sorts, and he accidentally forgot to bet on Leicester City in 2015. He bet on them every single year dating back nearly 30 years. And yet the one year that he forgets to bet, not only did Leicester City win, but the bet paid out 5,000:1.

Let’s step back and calculate the cost of that oversight for Mr. Micklethwait.

£20 * 5,000 = £100,000.

A hundred… thousand… pounds. That kind of winning would put a nice dent in your mortgage, wouldn’t it?

The risk of low probability strategies

Everyone hears anecdotes about successful trend traders. Even winning only 30-40% of the time, they walk away big winners over time.

You live HERE. Math isn’t good enough. You also need to wonder if your strategy can handle real-world problems.

What if they took that even lower? They could move their stop losses closer to the market. They’d reduce the size of the average loser, but the winning percentage might also drop to 10-20%.

Mathematically, this could work out identically. 30% winners that earn 5x the average loser make for a profit factor of 1.5. A strategy with only 10% winners that make 15x the typical loser also have a 1.5 profit factor.

They’re the same. Aren’t they?

Planet Earth isn’t the same as planet Math. In the real world, people get sick and miss trades. Or, they move across the Atlantic and forget to place a £20 bet.

People move. They get sick. Computers break. Things can and will go wrong with trading.

Richard Dennis once commented that the Turtle Traders would often make their annual returns off of one, single trade. A single trade!

When your performance depends on positive outliers, you’re massively vulnerable to accidents. What happens if you’re sick that day? Or your internet goes down? Or your broker locks you out of your account on the worst possible day?

Life happens, brother. A plan that depends on perfection is no plan at all. You need to make yourself robust to failure. Or even better, you’d make yourself antifragile.

Winning percentages

I mentioned that you can do really well winning 30-40% of time. Why then, does my own trading strategy, Dominari, win 68% of the time?

Because I’m exploiting compound, exponential growth. It’s not just how much you win, but the order in which you win it.

Let’s take two examples:

A ranging strategy with a profit factor of 1.3 that wins 68% of the time.
A trending strategy with a profit factor of 1.3 that wins 30% of the time.

Look at the red circles. Trending strategies are prone to extreme outcomes, both positive and negative.

Each strategy risks about 1% on any given trade. And, the average of the range and trend strategies are identical in the long run.

But… and this is an important “but”, the expected worst case scenario with the trending strategy is substantially more likely compared to the range trading strategy. In effect, the average is more average with a ranging strategy than with a trending strategy.

Why is that? Because unusual losing streaks are devastating to trending strategies. Have you ever had a losing streak? It happens to everyone.

By using a strategy with a higher winning percentage, you’re making yourself robust to streaks of losers. And, not to mention, your average length of a winning streak is considerably higher.

Even though you’re getting the same mathematical outcome, you’re making things much easier on your trading psychology when you adopt a strategy with a higher winning percentage.

Dominari & Exponential Growth

You may have thought to yourself, “68%? That’s kind of a strange number to pick.”

You’d be right. The choice of 68% winners was not a coincidence. It is, in fact, the win rate on my Dominari strategy.

Dominari is about more than just buying and selling. Trading is also about managing a portfolio and position sizing. Position sizing is phenomenally important over your trading career.

My backtest results for Dominari show that for every $2,500, the account increased to $17,855.35 after 3 years. That kind of compound growth doesn’t happen by accident. That’s why I’d like to share the good news with you in my webinar this week.

I’m going to show you how to put that exponential awesomeness to work in your trading account. Sound good? Click here to register for the FREE webinar.

How to be 99.999% accurate when your system is only 49% accurate

November 16, 2015 by Shaun Overton 9 Comments

Virtu Financial, the high frequency trading firm whose initial public offering of stock was caught in the unexpected firestorm that was the book “Flash Boys” by Michael Lewis, is reviving the IPO plan they shelved last year amid controversy, is seeking $100 million.

99.999 win percentage is an odd statistic

As a recent Securities and Exchange filing reveals, the company, operated by a litany of some of the exchange world’s top executives, boasts that out of 1,485 trading days it has only one losing day. This is the key statistics that left those familiar with algorithmic trading scratching their heads.

On the surface this 99.999 win percentage is a rather unworldly performance statistic in the world of algorithmic trading.

Virtu Financial is not a trend follower

The most popular managed futures strategy, trend following, has an average win percentage near 55 percent. Trend following might not be the best algorithmic strategy to compare to Virtu, however, as the firm claims in its S-1 that their trading “is designed to be non-directional, non-speculative and market neutral.” Micro-trend following and benefiting from market moves in one direction is a popular high frequency trading strategy, but based on their S-1 this is not the primary strategy.

This doesn’t explain the win percentage.

The highest win percentage of all managed futures strategies, near 75 percent, is short volatility, which is also the least popular strategy. While the strategy is known to win most of the time, the key statistic is to understand its small win size and large loss size. In managed futures the size of a trader’s wins can often be more important than how often they win. In the case of short volatility, while they win most of the time, when they lose they lose big – with an average loss size that is close to double that of a discretionary trading category, for example.

While risk in Virtu may exhibit strong downside volatility during crisis, much in the same way market crashes bankrupt many individual market makers in the golden days of the trading floor, comparing Virtu’s strategy to a short volatility strategy is inaccurate.

Perhaps the most applicable managed futures strategy to benchmark might be the relative value / spread arbitrage category. The spread-arb strategy has a high win percentage, near 60 percent, and it also has the best win size / loss size differential. The strategy works by buying one product and then selling a related product. The directional strategy works when a market environment of price relationship dislocation occurs.

While the fit isn’t perfect, nonetheless the most relevant managed futures strategy for which to compare Virtu is its direction-less, market neutral approach taken by certain spread-arb CTAs. The primary difference being Virtu doesn’t hold positions for directional profit. What they do, much like a short term trend follower, is take a position and then immediately lay off risk in a hedge. For instance, they may buy oil and then immediately hedge that position in another market and perhaps even using a derivatives product with different product specifications. In the olden days one could simply describe this as a “market making” strategy, but in the new school world of high frequency trading, separating two-sided liquidity providers from directional trend followers has oddly become more difficult.

99.999 percent daily win percentage overshadows 49 percent intra-day win percentage, highlighting the importance of win size

When comparing Virtu to known managed futures strategies, the 99.999 win percentage sticks out like a sore thumb – until you read the next punch line in the most recent S-1. That is when the firm reveals that its win percentage on an intra-day basis is 49 percent.

This puts the pieces of the puzzle together. The 99.999 percent win percentage needs to be considered in light of the 49 percent intra-day win percentage. This highlights the fact that Virtu, by logical default, is benefiting from size of win. Just like a trend follower, it isn’t always win percentage that matters most but size of win and controlling loss. This is likely the secret sauce inside Virtu’s success.

This article originally appeared on Virtual Walk and was authored by Mark Melin.

Have You Prepared For System Failure?

January 3, 2014 by Andrew Selby Leave a Comment

One of the misconceptions that many quantitative traders fall prey to is neglecting to consider that their strategy will eventually stop working. We are led to believe that once we develop and backtest a profitable strategy, we will simply be able to print money indefinitely. However, this is almost never the case.

Due to the unpredictable nature of financial markets, all systems and strategies will eventually fail. At the very least, they will need to be adjusted. This means that developing a trading strategy is an ongoing process, not a one-time project.

Eventual system failure is inevitable for all types of strategies. Are you prepared for it to happen to you?

Daniel Fernandez from Mechanical Forex wrote a post on this topic earlier this week. He suggests that the ability to detect system failure with as little pain as possible is a pivotally important aspect of Forex strategy development. He explains why all quantitative strategies are bound to fail eventually:

Eventual system failure – what we can call system death – is an inevitable consequence of an edge developed on a finite amount of information on a market with potentially infinite variations.

Detecting System Failure

Fernandez made some particularly interesting points about the process of detecting potential system failure. In order to detect that a strategy no longer working, a trader will most likely have to go through a difficult losing period.

Through extensive backtesting, walk-forward testing, and monte carlo simulations, a trader can establish parameters that describe a normal losing period for a given strategy. In order for that trader to determine system failure, they will have to trade through that normal losing period and then some.

The interesting concept that Fernandez brings up is that different strategies will have different conditions for those standard losing periods.

Low Win Ratio Strategies

Trading systems that are based on low winning percentages and high reward to risk ratios are expected to have long losing streaks. Therefore, it would take an exceptionally long losing streak to signal that system failure is possible.

Fernandez also adds that these types of strategies often rely on a few very profitable trades to make up for lots of small losses. That means that missing one key trade could result in a false signal that the system has failed.

High Win Ratio Strategies

Strategies based on high winning percentages and low reward to risk ratios pose the exact opposite problem. They experience much shorter losing streaks, so they are able to identify system failure much sooner.

Of course, the losses that these types of systems do take are often very large. While it might be a short string of losses that identifies the system failure, those losses are likely going to be incredibly painful.

Best Strategies for Detecting Failure

Fernandez concludes that the systems that provide the least painful means of detecting system failure are strategies with moderate winning percentages and reward to risk ratios.

He suggests that systems with reward to risk ratios around 1 to 1 and winning percentages just over 50% are able to signal failure in the best manner. These types of strategies can signal failure quickly, without crippling the buying power of an account.

Trading Frequency

The last topic that Fernandez mentions is the trading frequency of a strategy. Again, he suggests targeting a middle-of-the-road approach.

The fact that high-frequency trading systems can run through long losing streaks quickly might be seen as an advantage. Fernandez points out that this can be a double edge sword. Short term disruptions in market behavior can lead to false signals of system failure.

Fixed Fractional Money Management

April 10, 2012 by Shaun Overton 8 Comments

Fixed fractional money management changes the overall outcome of your trades. Remember that trading is the net outcome of several hundred trades. The power of money management comes into play after several hundred trades or more.

Trading totally at random with a 50% winning percentage and an R multiple of 1 yields no advantage, as I discussed last week in modeling money management. Remember that an R multiple is the average win to the average loss. Such a system poses neither an advantage or disadvantage. The average outcome should come out extremely close to the starting balance.

Fixed fractional money management stretches some portions of the bell curve and compresses other regions. Before we get into that, it’s important to remember what fixed fractional money management means. It’s a complicated name for a simple concept. It stands for the idea of risking a set percentage of the current account equity rather than the starting equity.

Most traders focus on risking a set dollar amount such as $1,000 on a given trade. This method updates that dollar figure after every single trade.

Consider an example where the account balance starts at $100,000 risking 1%. Both methods risk the same amount on the first trade, $1,000. The next trade, however, will yield a different risk amount. A win on the previous trade would increase the account equity to $101,000. One percent of a 101 grand is $1,010 of risk on the next trade. A whopping ten dollar change.

That may seem trivial. It is most certainly not over the long run.

Examples

Consider a trader that plays the coin toss game and has a system with the following characteristics:

He starts with a $100, 000 account balance
His R multiple is 1.0
He wins 50% of the time with no trading costs
He risks 1%

The absolute worst outcome of playing the coin toss with a fixed dollar risk of $1,000 is a loss of $46,000. Adding fixed fractional money management during that difficult drawdown improves the drawdown to a less substantial loss of $37,500. The worst drawdown goes from -46% to -37.5%. The method drags the absolute worst case scenario and pulls it closer to the average. When an unlucky, devastating drawdown kicks in, the technique reduces the losses that the trader experiences.

The best case scenario for fixed dollar risk is a $58,000 (58%) return. Adding money management to the system dramatically stretches the best case scenario further to the right. It improves to a $76,000 return (76%). The good times get a lot better without changing anything at all about the trading system. The method stretches positive returns away from the average. The trader walks away with more money in his pocket.

The natural instinct is to conclude that fixed fractional money management is the way to go. I agree. It improves the risk reward profile of a totally random strategy. Adding it to a real trading system should help control parameters that most traders consider critical like drawdowns and maximizing the return.

An important consequence of using fixed fractional money management, however, is that the odds of receiving a below average return increase somewhat. The coin toss game suffered a below average return 47% of the time. Applying fixed fractional money management increased the likelihood of a below average return to 53%. The effect is not all that much. Losing is more likely. But when it happens, the “loss” is so negligible that it can be thought of as breaking even.

Random numbers occasionally follow a seemingly non-random pattern such as loss-win-loss-win. When this occurs, the size of the trade on the losses is bigger than the trade size of the winners. Even if the winning percentage comes out at precisely 50%, those wins get slightly overshadowed by the losers. That micro effect of slightly larger losses than gains shows up as a slightly increased risk of not making as much money as expected.

Graphing all outcomes

$Fixed fractional curves$

Fixed fractional money management changes the shape of the returns curve. This example is exaggerated to highlight the effect

Red areas represent the losing outcomes while green areas represent the winners. Money management is really about maximizing the ratio of green area to red area. Random trades with no expectation of profit yield a bell curve, which appears on the left.

Fixed fractional moves the highest density of returns slightly to the left. Doing so creates the trivial disadvantage of a slightly increased risk of negligible loss. Importantly, the far left side (the worst case loser) gets dragged much closer to the average. The far right side (the best case winner), gets stretched much further from the average. The green area is now larger than the red area. The random system has a minor expectation of generating a positive return!