Drifters & Steamers — The Risers & Fallers Of Betting Markets

Risers Fallers Drifters Steamers Betting Exchange

Risers and Fallers exist in both financial and betting markets. In finance the Risers & Fallers show us the “winners” and “losers” respectively, over a period of time (e.g. the trading day).

The only significant difference between Risers & Fallers in finance to those in betting is that in the betting markets we want the price (the odds) to shorten or “steam”, as opposed to increase or “drift”.

 

Drifters & Steamers In Sports Trading


► The Objective

To complete a successful trade on the betting exchange you must Back odds that you can later Lay at a lower value. This is the basic “back high, lay low” philosophy. In this post I analyse the results of a trading strategy and demonstrate the importance of backing at a “high” price, before the odds shorten.

 

► Predicting market Movements

Its very difficult to guess whether the odds you Back are going to fall (or “steam”). It’s inevitable that you’ll sometimes get it wrong — i.e. the odds you Backed didn’t fall to a value you wanted, or they didn’t even fall at all.

For this reason many traders only look to capitalise on small movements in the market and don’t worry about the long-term value of odds. These “short-term” traders profit from a small rise or fall in the price. This type of trading approach is practised by volatility traders, and “scalpers“.

Other traders take a long-term view and look to capitalise on large moves in the market. Big Fallers are colloquially known as “steamers”, and are exactly what long-term traders try to identify and Back at an early, high price.

 

► Analysing Risers & Fallers

What follows is a ‘mock’ strategy that identifies potential value, places the bets, and holds the position without trading out — regardless of market movements.

I used £10 bets on selected horses in the Win market. At the time of betting I believed the odds to have a good chance of falling (steaming).

  • 59,281 Back bets were placed on the betting exchange.
  • I used the SP odds as the point where I define whether the horse were a long-term “Riser” or “Faller”.
  • I assume 5% commission was taken by the betting exchange.
  • Bets were placed on the day of the race, at any point up until the off.
  • The average decimal odds for the bets was 8.66.

 

The Fallers — “Steamers”


The outcome of the Fallers will come as no surprise to most. But readers who hold a pessimistic view on the accuracy of odds may find it eye opening to see such a strong trend in the positive direction.

 

All bets with odds that fell to a lower value by SP

Risers Fallers Betting Exchange

The strategy had 31,448 Fallers which produced a yield (ROI) of +19.62% after commission. This accounted for 53.04% of the total bets placed.

It’s actually a pretty good sign that the strategy Backed more horses above the SP odds than below it. The odds Backed were an average of 13.72% above the SP odds.

This graph’s strong positive trend precisely illustrates the value in achieving good early price on horses before the odds eventually settled at a lower value. But this alone does not mean the selection process is profitable; it ultimately depends whether the positive yield seen in the Fallers outweighs the yield on the Risers.

So let’s see…

 

The Risers — “Drifters”


The result of the Risers is as consistent and clear-cut as the Fallers, albeit in the opposite direction.

 

All bets with odds that rose to a higher value by SP

RISERS

In this strategy there were 25,693 Risers, accounting for 43.34% of the total number of bets, producing a negative yield of -22.35% after commission.

The strong negative trend (which is seemingly a reflection of the Fallers), underlines the the importance of:

  • Selecting bets which go on to Fall (Steam), and
  • Closing out of bets that appear to be moving against you (i.e it’s rising). Or avoiding those bets altogether!

Recall that no closing out process was used in this “Back only” strategy. So no action was taken to lessen the impact of rising odds. Hence the deleterious effect these Risers (drifters) had on the overall profit.

The average closing SP odds for the Risers was 20.16% above the odds Backed by the strategy. Given that the SP represents approximately fair value at the time race begins, it’s no wonder the Risers lost money!

 

Non-Movers


This third and final category has less occurrences — but due to the large data set in our test strategy I’m able to make relevant observations.

 

All bets with odds that remained the same by SP

NONMOVERS

There were 2,140 Non-movers, accounting for 3.6% of the total number of bets, producing a yield of 1.09% after commission.

Non-movers belong in a kind of “Purgatory”. The odds haven’t moved significantly enough to be classified as either a drifter or steamer.

The odds which didn’t move, or otherwise reverted back to the same odds, represent occurrences where the strategy obtained approximately fair value. So our prediction of the odds falling was neither right or wrong.

The results lack consistency as shown by a graph that gives no strong indication as to what direction the trend is heading.

 

The Combined Result — Fallers, Risers & NonMovers


With all three groups, Risers/Fallers/Non-movers combined, the yield remained positive at 2.29% after commission.

 

All bets combined

COMBINED

Whilst the Risers carry a large negative yield of -22.35%, the Fallers’ positive yield of 19.62% and the fortunate result of the NonMovers at +1.09% is enough to produce a positive result overall after commission.

What proved influential to the positive overall trend was the fact that the Fallers counted for 53.04% of the total number of bets, outweighing the Risers which counted for only 43.34%.

Interestingly the Risers drifted comparatively further away (+20.16% move) from the odds we backed than the Fallers (-13.72% move). This raises doubt that the Fallers carry enough value for the profit to outweigh the further-drifting Risers. However, the difference in the two yields (in this case) opposed one another in a relatively equal fashion, with only -2.73% additional negative yield attached to the Risers than the positive (Fallers). So the greater-sized negative yield wasn’t enough to make a detrimental impact to the overall result, as shown by the graph.

With regards to the test strategy — or indeed any betting strategy — it’s useful to monitor the movement of the market prior to the race starting. This gives you a good idea of how valuable your odds really are. In general:

  • If odds Fall a significant amount by the SP then you have positive value
  • If odds stay still, you have fair value, and
  • If odds Rise significantly, then you have negative value.

There are of course occasions where the market gets it wrong, or the odds are inaccurate. This presents an opportunity for traders to snap up high odds with Back bets, or Lay where the odds have been over-exaggerated and “steamed” too far.

 

Improvements


The mock strategy isn’t an optimal approach. I have simply demonstrated the outcome of holding bets and proved that the movement of the odds correlated to how successful the bets were.

We can improve this strategy by applying two rules:

  1. Neutralise the Risers by trading out of a bet. This means breaking even with a “scratch trade”, or accepting a small loss.
  2. Preserve the Fallers by holding onto a position till a later point, or simply holding the position without Laying off.

Applying those rules is a good way to trade in general. Still, it’s easier said than done. It takes a lot of trial and improvement as well as experience to improve your trading strategy to the point that it makes consistent profits.

Be aware that prices sometimes Rise initially and then go on to Fall heavily. Other times prices Fall first and then Rise heavily. It’s not always as clear cut as it seems, so I recommend using as much data available to you whilst you trade.

You may better predict market movements using methods such as the Weight of Money. Geeks Toy or BetTrader have an abundance of metrics designed to help traders make informed decisions. Without specialist Betfair trading tools or bots you’re already at a disadvantage in what has become a very competitive market.

Perhaps the most important point to take away from this elaborate demonstration of Drifters & Steamers is that the betting exchange is pretty efficient. The results of thousands of bets are testament to that. So let the market tell you what bets you need to keep and those you need to trade out of.

 


Further Reading:

The Accuracy & Significance Of SP Odds

What Causes Pre-race Market Movements?

Cold Trading On The Exchange — Betting Without Knowledge

  • Tom Davies

    Good demonstration, and it makes sense. “Back high lay low” makes money for a reason – “back high” to grab a good price, and holding it (without laying off) is value… if you assume the market has been ‘right’ in adjusting the price down to a lower value by ‘the off’. Fascinating to see how consistent the trend is even with such high odds/risk. So how did you collect so much data for this??

    • At the time of writing I was using the Betfair API to collect data. I no longer scrape/collect odds.
      Before checking the results, I actually didn’t expect the trends for the Drifters & Steamers (respectively) to be so smooth. It’s was an eye-opener.

  • Rob Pitcher

    I’ve read in forums that beating the SP isnt consistent enough. It doesn’t seem that way based on these results.

    • Those opinions are most likely based on a small set of data.

  • Ivette T. Davis

    Is it easy enough to trade out of something that’s starting to drift and then keep those ones which steamed? It looks like can profit so long as you scrap the ones that are moving against you.

    • That’s the right idea – but it’s harder in practice than it seems. Sometimes the odds look like they’re moving against you and start to drift… and then they’ll suddenly steam afterwards. Often something steams to begin with, and then drifts. This regularly happens late on, right before the start of the race.

      So in theory it sounds easy to simply “cut your losses” on the drifters, but it’s hard to single them out with accuracy.