I have always been diligent about looking at past performance of mutual funds prior to investing, and consider it to be one factor when comparing mutual funds. However a recent improvement I made to the VFS Daily Investment contest caused me to look at historical performance in a different light.
Starting in 2016, I am providing a prize for the portfolio with the best 3 year cumulative return. The purpose was to reward ‘long-term’ excellence, and assumed the top portfolios would be pretty consistent. However, I was surprised when I saw the results for 01/31/16 – the rankings for the top 10 portfolios changed dramatically:

So this caused me to immediately assume my calculations were messed up – so I did a little analysis of the top two portfolios:
Karen’s Portfolio |
Greg’s Portfolio |
|
Month |
Monthly Return |
$100 invested
1/1/13 to 12/31/15 |
$100 invested
/1/13 to 1/31/16 |
Monthly Return |
$100 invested
1/1/13 to 12/31/15 |
$100 invested
2/1/13 to 1/31/16 |
1/31/2016 |
-3.89% |
|
$ 190.29 |
|
-7.54% |
|
$ 181.96 |
12/31/2015 |
2.34% |
$ 197.24 |
$ 197.99 |
|
-4.98% |
$ 206.56 |
$ 196.79 |
11/30/2015 |
2.61% |
$ 192.73 |
$ 193.46 |
|
5.99% |
$ 217.37 |
$ 207.09 |
10/31/2015 |
19.97% |
$ 187.82 |
$ 188.54 |
|
9.37% |
$ 205.08 |
$ 195.38 |
9/30/2015 |
-2.58% |
$ 156.55 |
$ 157.15 |
|
-4.11% |
$ 187.51 |
$ 178.64 |
8/31/2015 |
-2.06% |
$ 160.70 |
$ 161.31 |
|
5.02% |
$ 195.56 |
$ 186.31 |
7/31/2015 |
8.10% |
$ 164.07 |
$ 164.70 |
|
0.67% |
$ 186.21 |
$ 177.40 |
6/30/2015 |
-4.63% |
$ 151.78 |
$ 152.36 |
|
-1.96% |
$ 184.98 |
$ 176.23 |
5/31/2015 |
1.59% |
$ 159.16 |
$ 159.76 |
|
4.19% |
$ 188.68 |
$ 179.76 |
4/30/2015 |
1.75% |
$ 156.67 |
$ 157.26 |
|
5.40% |
$ 181.09 |
$ 172.53 |
3/31/2015 |
-2.02% |
$ 153.98 |
$ 154.56 |
|
-0.89% |
$ 171.82 |
$ 163.70 |
2/28/2015 |
7.40% |
$ 157.15 |
$ 157.75 |
|
5.35% |
$ 173.36 |
$ 165.16 |
1/31/2015 |
-1.26% |
$ 146.32 |
$ 146.88 |
|
7.62% |
$ 164.55 |
$ 156.77 |
12/31/2014 |
-3.65% |
$ 148.19 |
$ 148.75 |
|
-2.65% |
$ 152.90 |
$ 145.68 |
11/30/2014 |
2.77% |
$ 153.80 |
$ 154.39 |
|
2.32% |
$ 157.06 |
$ 149.64 |
10/31/2014 |
5.12% |
$ 149.65 |
$ 150.22 |
|
0.23% |
$ 153.50 |
$ 146.24 |
9/30/2014 |
-2.45% |
$ 142.36 |
$ 142.90 |
|
-1.79% |
$ 153.15 |
$ 145.91 |
8/31/2014 |
1.93% |
$ 145.94 |
$ 146.49 |
|
4.60% |
$ 155.94 |
$ 148.57 |
7/31/2014 |
1.37% |
$ 143.17 |
$ 143.71 |
|
-1.79% |
$ 149.08 |
$ 142.03 |
6/30/2014 |
0.65% |
$ 141.24 |
$ 141.77 |
|
3.03% |
$ 151.80 |
$ 144.62 |
5/31/2014 |
5.11% |
$ 140.32 |
$ 140.85 |
|
6.09% |
$ 147.34 |
$ 140.37 |
4/30/2014 |
-0.33% |
$ 133.50 |
$ 134.01 |
|
-0.15% |
$ 138.87 |
$ 132.31 |
3/31/2014 |
-0.98% |
$ 133.95 |
$ 134.46 |
|
-3.44% |
$ 139.09 |
$ 132.51 |
2/28/2014 |
6.70% |
$ 135.27 |
$ 135.78 |
|
4.00% |
$ 144.04 |
$ 137.23 |
1/31/2014 |
-2.99% |
$ 126.78 |
$ 127.26 |
|
-1.80% |
$ 138.50 |
$ 131.95 |
12/31/2013 |
2.63% |
$ 130.69 |
$ 131.18 |
|
4.74% |
$ 141.04 |
$ 134.37 |
11/30/2013 |
2.86% |
$ 127.34 |
$ 127.82 |
|
3.31% |
$ 134.65 |
$ 128.29 |
10/31/2013 |
9.14% |
$ 123.80 |
$ 124.27 |
|
7.58% |
$ 130.34 |
$ 124.18 |
9/30/2013 |
0.90% |
$ 113.43 |
$ 113.86 |
|
5.08% |
$ 121.16 |
$ 115.44 |
8/31/2013 |
-1.22% |
$ 112.42 |
$ 112.84 |
|
-1.51% |
$ 115.31 |
$ 109.86 |
7/31/2013 |
4.78% |
$ 113.81 |
$ 114.24 |
|
9.06% |
$ 117.07 |
$ 111.54 |
6/30/2013 |
-1.71% |
$ 108.61 |
$ 109.02 |
|
0.24% |
$ 107.35 |
$ 102.27 |
5/31/2013 |
1.48% |
$ 110.50 |
$ 110.92 |
|
1.08% |
$ 107.09 |
$ 102.02 |
4/30/2013 |
3.26% |
$ 108.89 |
$ 109.30 |
|
1.44% |
$ 105.94 |
$ 100.93 |
3/31/2013 |
2.90% |
$ 105.45 |
$ 105.85 |
|
2.15% |
$ 104.44 |
$ 99.50 |
2/28/2013 |
2.87% |
$ 102.48 |
$ 102.87 |
|
-2.60% |
$ 102.23 |
$ 97.40 |
1/31/2013 |
-0.38% |
$ 99.62 |
|
|
4.96% |
$ 104.96 |
|
And I found my explanation. Using the approach of measuring return in terms of ‘$100 invested 3 years ago now is worth x’ does really magnify recent performance. Note that even though Gregs portfolio was down 7.5% in January, it knocked $15 off his hypothetical investment. So losses for the leaders are magnified as they hypothetically have more assets to lose.
The other important point is the starting point makes a big difference. Greg’s portfolio for the period 2/1/13 – 01/31/16 got hurt, because he had a decent month in January (highlighted in green), which no longer counted. Removing the 4% gain from the first month knocked $10 off the 3 year return, further causing the large drop in Greg’s portfolio’s value.
So the moral to the story? Be careful when looking at mutual fund past performance – as the time period you are looking at may make a big difference in how the performance is measured. Or better yet – maybe mutual fund expense ratio is a much better indicator of future fund performance, since past performance is a pretty fluid measurement.