It’s time for another quarterly review. The last few months since the last review have followed on in a similar vein  i.e. volatile and only marginally ahead of the FTSE. I would expect my strategy to perform less well in more volatile periods, so being marginally ahead of the FTSE is a small comfort. Though as always the point of the review is be critical and look for ways to improve.

For ease of reference, here is a link to my portfolio so you can see what’s in it at the moment. It’s actually a fantasy portfolio set up on Stockopedia to track my actual portfolio. The overall value is different but I try to keep the proportions of each holding as accurate as possible so it can track my actual performance. The fantasy portfolio excludes dividends and tends to buy at slightly worse prices than I can get from my broker, so it somewhat underestimates the performance of my actual portfolio.

Performance

Benchmarks

The performance of my portfolio (QSS) is shown against its benchmarks in the table below (all the figures below exclude dividends).

My benchmarks are:

• The FTSE100
• The top ranking decile of stocks according to Stockranks
• A ‘buy and hold’ benchmark portfolio of the 100 shares on my portfolio and watchlist

I’ve dropped the mechanical benchmark portfolios as these were based on the ranking in my spreadsheet. It was proving too much hassle to keep these up to date. They have been replaced with a more straightforward momentum following benchmark portfolio, but it is too early to report on it.

While over the longer term I am well ahead of all the benchmarks, my recent performance has not been so great. I am ahead of the FTSE100 over the last six months, but this is not by much and I’ve fallen behind the top Stockranks stocks over this period.

The strong recent performance of my ‘buy and hold’ benchmark portfolio is reassuring in that it suggests my focus on high quality shares continues to work well. However, I am starting to get concerned that my actual portfolio has recently been underperforming it. This suggests that my trading decisions have not been adding value over sitting still. This is partly to be expected as market volatility has increased, undermining the benefits of momentum following. As I explore further below, I don’t think I’ve dealt with the increased volatility very well. Hopefully I can improve my strategy and get back on track.

Trading stats

The table below shows my trading statistics. The key things to note are the win/loss ratio, which shows the proportion of profitable trades, and the gain/pain ratio which shows the size of an average win relative to an average loss. What I am aiming for in my momentum based strategy is to have a win loss ratio above 40% but to ensure the gain/pain ratio is as high as possible. I do this by cutting losers before they are more than 10% down and by holding and adding to winners.

Unsurprisingly my trading stats have been declining along with my performance. My win loss ratio has fallen below 40% which is a bit troubling. I think it must suggest that I am overtrading: either I have been buying too often when the reward to risk is poor or I have been cutting my losers too quickly. Thinking back, one thing I haven’t done very well is keeping a cash balance when the volatility amps up, despite intentions to do so. I might benefit from formalising this rule to give myself more discipline. It also looks like I could have benefitted from a looser stop loss policy. I’ve looked at this in more detail in the context of individual trades below.

My gain to pain ratio has also fallen, though is still at a respectable level. The size of both my average gain and my average loss has risen. This makes sense – as my portfolio has grown so has the size of my trades. However, the size of gains is rising a bit more slowly. This suggests to me that I may be cutting winners too quickly. Thinking back, I have been applying trailing stop losses to winning as well as losing positions – perhaps I’d be better served by allowing my winning positions more room to oscillate as long as I am in profit? I’ve also looked at this in more detail below.

Strategy review

Over-trading analysis

My main concern going in to this review was that I have been overtrading. My performance stats are consistent with this, but to properly check I need to look a bit more granularly at individual trades.

My starting point is to revisit my ‘overtrading’ analysis. This looks at what happened to shares after I sold them from my portfolio: did they go on to do better or worse than the rest of the portfolio? If the shares I’m selling are generally going on to outperform my portfolio this indicates I must be over trading. I’ve looked at this two times previously: in April last year I looked at what had happened to the shares I sold over the latter half of 2016, and in September I looked at the first half of 2017.

Both of the previous times I looked at this, the evidence was pretty compelling that I was not overtrading. For the 2nd half of 2016, only 4 out of 29 shares I sold went on to perform within 5% of the rest of my portfolio. For the first half of 2017, only 7 out of 36 did so. While the shares I was selling were not doing so badly, the rest of my portfolio was doing so well that the opportunity cost of holding even moderately performing shares was very high.

I’ve repeated this analysis for the second half of 2017. My overtrading analysis lags the rest of my performance stats as I need a bit more time to see what would have happened to the shares I sold. The results are a bit shocking – this time 27 out of the 34 shares I sold have gone on to do better than (or within 5%) of the rest of my portfolio and I’ve missed out on a few really big winners: AB Dynamics, NMC Health and Accesso Technologies have all risen around 50% or more since I sold them. What was working well before seems to have dramatically reversed as volatility has risen.

To look at this in more detail, I distinguished between those positions I closed for a profit and those I closed for a loss. Of the 14 trades I closed for a profit, 13 went on to perform better than (or within 5% of) the rest of my portfolio. This clearly suggests that I am cutting winners to quickly. Looking back it seems most of these sales have been made because of falling momentum. It seems I need to give my winners more room for their share prices to oscillate when conditions are more volatile.

Of the 20 trades I closed for a loss, 14 went on to outperform the portfolio. This suggests I have been also cutting losers too quickly, but not to the same extent as the winners. For these trades I’ve considered what would have happened had I applied a looser stop loss policy. It looks like most of the shares I sold with a 10% stop loss would have also been sold even with a bigger 15% stop loss, so this generally wouldn’t have helped. However, there are a small number that I cut before 10%. Had I avoided doing this I would have netted a couple of additional big winners – Accesso and AB Dynamics.

Refining strategy

I need to take the results of all this analysis with a healthy pinch of salt. Chance plays a significant role – perhaps I was very lucky to hit so many big winners over the last couple of years and had I had the better judgment to hold on to only a couple of the big winners I sold this year, this could have dramatically affected the results. However, to a large extent I think it’s sensible to take the results at face value when considering how to modify my strategy and pick up any nuances in the next review.

The overall lesson I think is that I need to adapt my strategy to market conditions. Tight stop losses and close momentum following can work spectacularly well in a low volatility environment. It may even work well across the whole market cycle, after taking account of the temporary underperformance when volatility rises. I need to be careful not to throw the baby out with the bath water. However, it seems that I can do even better by adapting the strategy when volatility rises.

For this to work, obviously it’s important to be able to tell what market conditions are at any given time. Otherwise it’s all a bit circular. Luckily, I don’t think this is so difficult. For a start I’m more interested in market conditions for the kind of shares I invest in rather than the market as a whole. That means I can learn a lot from simply looking at how my portfolio is doing. When performance starts to dip, say by more than 5% from its high, this indicates that conditions have become more volatile and I should adjust my strategy. The other thing I think is worth monitoring is the number of shares in my buy and hold portfolio that are near their highs – this can tell me whether the volatility is being experienced by my whole watchlist or just the ones I’m holding in my portfolio at the time. I can get at this by monitoring how many shares are within 5% (currently 18) and 10% (currently 45) of their 52 week highs. I’d say current conditions were mixed to bearish, giving me a rough benchmark for what this should look like when conditions are improving.

When conditions are less volatile I feel I can assume what I’ve been doing works well i.e. I should use tight trailing stop losses to cut positions quickly to transfer the funds to shares with better momentum, especially where there is also positive news flow.

When my portfolio and watchlist start to dip and market conditions have become more volatile, I think I should do the following, based on my experience over the last six months:

• Don’t take profits on winning shares. While I may want to raise some cash, I should look to the losers for this. I want to hold on to my winners through the volatility (unless there is a change in the underlying story of course).
• Relax stop losses and allow a bit more volatility. I don’t want to be trading in and out of positions so much and need to give a bit of room for prices to oscillate. I should only sell where I have significantly less conviction than holding cash or an alternative investment, rather than simply if momentum is weakening.
• Rotate towards higher conviction defensive shares. To give myself the confidence to hold through volatility, I want to rotate towards shares where I have greater conviction. While everything in my watchlist is high quality, in practice this means holding more defensive low volatility shares with less correlation to the overall market and where valuations are not stretched.
• Hold on to some cash to hedge against a bigger fall. Having some cash on the sidelines to take advantage of a crash is an important insurance policy that also takes some of the emotional sting out of a falling portfolio. Over the past few months I’ve been thinking this is a good idea but haven’t had the discipline to actually implement it, so I think I need a rule. I think this should be to maintain at least 20% cash until conditions improve. I don’t have to necessarily raise the cash balance immediately, but I should bar myself from making a new purchases that would reduce this cash balance below 20%. I can raise the cash over time by selling out of lower conviction holdings that hit my stop losses.
• If conditions worsen to a fully-blown bear market, focus on planning how and when to reinvest the 20% cash balance. I want to make the most of my cash balance ‘insurance policy’. Broadly this means holding off until the market has fallen quite a bit and then waiting for a point of maximum pessimism – a big down day, following a sequence of down days. Patience is key – if the market hasn’t yet fallen that far then I should keep the insurance – if I don’t need to use it then fine.

These steps are designed to move my portfolio towards a more defensive one with a significant cash balance that I have the confidence in to hold through the volatility rather than try to trade it. I should only move back to my close momentum following strategy  if the volatility subsides and my portfolio is steadily making new highs again or if the market has crashed and is starting to rebound.

I intend to start applying this immediately. The implications of these rules for my current portfolio, which is currently facing volatile conditions, are broadly that I should a) hold off from any new purchases until I have raised my cash balance to 20% (it is currently at 7%) or conditions have improved, and b) only sell out of lower conviction losing shares until my cash balance hits 20%. Everything I hold is fairly high conviction at the moment, so I’m likely to hold on unless conditions start looking very bearish.