Forum Topics Too Much Information!
PortfolioPlus
7 months ago

I’ll fess up to it – I am an information ‘wonk.’ In my business days, information was a ‘currency,’ and I made a reasonable coin out of it. Today, information is everywhere, and its largely free.

When it comes to investing, I look at so many individual ‘monitors,’ which takes time, but it scarcely improves my batting average; yet I feel very confident every time I stride out to the investment crease.

So, I went searching for a reason in the realm of psychology and I came across an experiment by Paul Slovic back in 1974 which has been repeated a number of times subsequently and found to be correct.

Short cut to the conclusion: Whilst its logical to think being ‘better informed’ leads to better decision-making, the flip side of the coin might well be that the additional information could just be topping up on confirmation bias, as we strive to prove ourselves right.

Here’s the experiment and the conclusions represented graphically.

In 1974, Slovic gathered a group of experienced horse racing handicappers to assess their success in predicting the outcome of 40 horse races conducted over four rounds. Because there were exactly 10 horses in each race, each handicapper’s bet could be expected to be right 10% of the time through random guessing alone.

Round 1: He gave them just five pieces of information they wanted on each horse such as age, fastest speed, jockey’s experience, or weight etc The result was that they were 17% accurate (fair enough as it is nearly twice the random result). He asked them to judge their ‘confidence level’ in their selection which was which was 19%.

Round 2: they received 10 pieces of information

Round 3: 20 pieces of information

Round 4: 40 pieces of information.

Now, despite the additional information, it didn’t improve their accuracy. It remained at 17%. But their confidence increased the more information they received to 34%.

0ade3931a98a3619adb28b05810b95dbacb96f.png 

So, the real question to be answered is this: Am I really becoming more accurate through grabbing more data or am I just fooling myself to prove my ‘want to believe’ thesis is correct? And does this additional deep dive on data make me anchor to my original thesis going forward?

In many instances, I am guilty as charged! Thankfully, I have evaded sharing a cell with Big Bubba with all the painful consequences…just a damn good thrashing with swimming noodles!

In 2024, I am going to rationalise my informational requirements and listen more to the ‘thoughts of the market’, which I previously viewed as Hoodoo Voodoo!

31

Solvetheriddle
7 months ago

@PortfolioPlus i think the above makes sense. i try and construct a bear case on all my large holdings. if I cant find one, it doesn't make me more confident it makes me more nervous! lol

14

UlladullaDave
7 months ago

Here is an ETF that combines a bunch of fund managers' highest conviction positions. A real life example of confidence/knowledge bias. James Montier's book "Behavourial Investing" has knowledge as gluttony in his 7 deadly sins. I'm not sure if the book is still in print. It's worth a read if you can find a copy.




1ab42e336579238c5fa95488ddf707ef0a3631.png

11

SaberX
7 months ago

I guess this is the old trade-off balance between time spent, and being 'nearly there' but getting the majority of results despite putting in less efforts.

It reminds me of work and in personal life. One could spend hours and hours perfecting and getting something perfect, receive no feedback or comments/issues, and then do something in 1-2 hours and either potentially get feedback and issues, or none at all (in which case you've saved yourself hours and hours of extra work you would have otherwise done).


I've found it's almost the same as information paralysis analysis, the old 80/20 rule etc. etc. I guess throughout time the same holds true as it always has for hundreds of years. Adding in sound rules and guidelines to your trading or investing plan I suppose minimises the risks as much as possible and introduces hopefully a player's edge, so that less time spent and trying to get prefect accuracy, doesn't bite you in the bum when it goes wrong.


At least that's what i'm hopeful of refining and working towards. It's the old DCF model issue - trying to model a 100 things and as Andrew put it in a recent podcast (well perhaps not recent, but me catching up on backlog of podcasts) trying to work out the R & D team's budget in a spreadsheet. Being generalist enough to have high level macro approaches to rationalising things, and getting a feel for value in general/at a much higher level. probably not the best quote of Andrew's words and in turn his reference to BUffet's approach i.e. a lack of a DCF excel spreadsheet, but you get the gist!

12

SaberX
7 months ago

@UlladullaDave - Love the concept behind it and it's purpose, would have loved to see a good graph on it.

5