Experiment HN: Crowdsourcing Tech Stock Predictions

http://www.zacharyburt.com/hn-crowdsourcing-stocks/

by zackattack • 16 years ago

32 35 16 years ago

puredemo• 16 years ago

Make sure to tell me how I did in six months. ;p

zackattack9• 16 years ago

I'll post a follow-up, but I'm not going to release people's individual predictions publicly in case I inadvertently embarrass anyone. I'm probably going to contact the winners individually to alert them of their insight and luck.

But yeah, I'll post a detailed analysis in 6 months' time. If people then want to contact me directly to get their data then they will be able to.

slashcom• 16 years ago

You should post aggregate values in a few days as well.

I hope you're also storing timestamps with the submissions. Just guessing, but I'd bet any predictions made on 04/30 will correlate much better than those made today.

xentronium• 16 years ago

Yeah, I second that and provide any prediction I make to the public domain!

joshu6• 16 years ago

The stock market is already a crowdsourcing platform. As is the options market.

zackattack9• 16 years ago

The interesting thing here is to see whether a community of opinionated elite nerds can do a better job of predicting the future direction than the general public / analysis / whoever. Speaking of which, have you logged your predictions yet?

joshu6• 16 years ago

Yes. By buying positions.

leelin2• 16 years ago

True, but the aggregation function is a bit complex.

I'm curious where we'd wind up with something more transparent or democratic.

Sadly, we aren't able to make any beta-neutral or relative value bets!

joshu6• 16 years ago

In what way do you think the aggregation function is "complex" or that it is not transparent or democratic?

You are stating opinions as facts. Please back them up?

keyle2• 16 years ago

It's more an experiment on chaos theory and dart throwing than "Predictions". I recommend you look at the fundamentals and aggregate the EPS forecasts in relation to the ROE to get a better idea as to where they should be in a year's time. On top of that you need to factor your economy and China's.

zackattack9• 16 years ago

You seem to know what you're talking about. I hope you submit a prediction. I doubt you will, though.

keyle2• 16 years ago

You're right I won't. It's like trying to predict the next lotto numbers.

Fear and greed fuel the market in the short term, so even if you got close 2 days before the deadline, a scare in the market would throw off all your predictions out the window.

All you can really predict is what you believe a stock will be worth (intrinsic value) assuming the EPS forecast is fairly correct, the economy behaves ok and that china doesn't cock up.

edit: I do appreciate anybody taking time to run an experiment though! That's why I voted up your post.

drcode• 16 years ago

Just to be fair to everyone entering things into this form: After you're done, you don't get to see any aggregated predictions, only a "thank you" message (which I found disappointing)

zackattack9• 16 years ago

Fair enough.

I would appreciate it if nobody shares their predictions in this thread, either. I know this isn't at all rigorous, but I'd still like to try.

kevinburke• 16 years ago

Does everyone get an equal vote?

trotsky• 16 years ago

SELECT username, MAX(one_year) AS price_target FROM guesses GROUP BY ticker;

A cunning strategy to discover Mary Meeker's HN account name.

strooltz• 16 years ago

are you planning on removing blank entries (those set at 0.00). i, for one, only filled out predictions who i actively follow in the market to keep "uneducated" data from entering your prediction. maybe a good strategy would be to remove the lower and upper ends of the curve as well...

zackattack9• 16 years ago

thanks, yes, i am planning on tossing out obviously bad data.

nandemo• 16 years ago

However, 0.00 might be a valid guess. If I think Acme is going bankrupt soon, then 0.00 is a good approximation for its future stock price.

jbooth2• 16 years ago

These are mature stocks, which are basically a random walk, unless you have some insider info.

What'd be really interesting would be if you got a couple thousand speculative plays on there, and allowed people to only submit on the industries they knew something about. Although that'd probably just lead to people only submitting their pet stock as "going to the moon!"

Nice job in any case though.

c1utch• 16 years ago

Every statement you made is wrong.

1. Every stock is a random walk. 2. Insider info for smaller companies is much more valuable then larger companies. 3. Speculative plays would make this even more boring, because then everyone makes wild guesses at the future value of a company they have never heard of.

Nice job in any case though.

jbooth2• 16 years ago

Every statement you made was a complete misunderstanding of my statement. Including the sarcasm, I was being genuine in my praise.

Briefly, the random walk factor is less of an issue for speculative stocks because they either make it as a business, or don't. That's not a random walk. And I really don't understand why you're so miffed.

ary• 16 years ago

Wasn't there just a story linked here the other day about the advice given to the newly minted (back at the IPO) Google millionaires? As I recall it said;

Don't try to beat the market. Don't try to beat the market. Don't try to beat the market.

larrykubin• 16 years ago

Some other popular sites with similar ideas are caps.fool.com and YC funded socialpicks.com.

aantix• 16 years ago

The biggest issue with this design is that there's no incentive to put in effort for better predictions.

Take it to the next level; design this for Mechanical Turk (https://github.com/aantix/turkee), and make this offer; for each prediction received, get a nickel. If you're on the right side with your prediction, receive an extra nickel. If you're within 0.5%, receive an additional nickel.

Not enough people utilize the "bonus" feature of Mechanical Turk as financial incentive to increase the quality of data they receive. You don't have to pay everyone the same. Reward better input.

zackattack9• 16 years ago

the point is to see if the HN crowd is good at predicting, not random people...

Groxx• 16 years ago

Somehow I don't think MSFT is $0.00.

Logged, with wild randomness. I don't pretend to understand chaos theory nor astrology.

Experiment HN: Crowdsourcing Tech Stock Predictions

Related Stories