I've been playing with stock data and I've discovered a terrible truth : in 10 years apple stock price passed from roughly 3\$ to 100$. However this gain isn't due to intraday price variation dayN_close - dayN_open
but the overnight variation dayN_open - dayN-1_close
.
In fact when you sum up the intraday variation
since 10years you get roughly -48$
drop in price (starting at 3\$ in 2004) while when you sum up the overnight variation
you get +148$
augmentation in price (148-48=100$ nowadays).
This mean that the reason why apple stock price went from 3 to 100 in 10years is the overnight variation in price. This is quite unexpected, if there was no overnight variation the stock price would have died a long time ago... Why is that ? Why do they say that intraday traders close their position at then end of day while most gains can be done overnight (buy just before the market close and sell just after it opens). Is this observation true for other symbols too or is it specific to apple ?
Edit : Well I performed some additional analysis and it turns out that the gain due to overnight variation is because over 10 years the frequency of the o/n variation being positive is by far superior to it being negative. As for the intraday variation the probability of it going downwards is slightly superior to it going upwards. However the variations that are high (>3% let's say) are more frequent intradaily. o/n variation tends to be small but steadily positive basically. As for the data I took it from nasdaq.com so I hope it isn't glitchy.