no one know this, and this is way i'm taking into account all possible scenario, speculation if done right must be done, by taking into account every possible future of what you're speculating about
there are not more chances that the demand will go down instead of going up for example, same ratio, simply because those variables are unknown, you can't say that it will go down for sure like i can't say that it will go up for sure
but what we know is, if you do not consider the demand because it is unknown, thus you can't consider it, if it is unknown, it is evident that for every halving the price will rise, it is really a simple concept
You aren't taking into account all possible scenarios, you are taking a default that the demand WON'T change, and that every other variable WILL remain constant. Which is completely silly because that's impossible. We don't know whether demand will go up or down, it's probably a 50/50. But we do know that it will NOT remain constant or near constant, that's just not possible. So to assume that as the default is silly.
Any change in demand will far out shadow the effects of mining rewards.
People don't get that it doesn't really matter whether the coins are kept, sold on Bitfinex, sold privately, or whatever. The bitcoin economy as a whole has to absorb that reward in any case.
Someone has to buy the coins. If the miner keeps them, then he himself is buying the coins. We can't pretend those coins aren't sucking something out of the BTC economy because if there were no block reward, he would have to buy existing coins to maintain his BTC income.
After all, there are over 10 million bitcoins, to think that a few thousand coins can change the price by thousands of percents is just silly.
Thousands of bitcoins do change the price in an "all else being equal" scenario. In a situation where we
would be going sideways with equal buying and selling interest, the block reward breaks the tie and sends us lower.
No. What YOU'RE not getting, is that there is money both going IN to bitcoin, and going OUT of bitcoin. For example, if everyday early adopters or people who lost faith in bitcoin after buying at the peak decide to completely exit and never come back, and sell $500,000,000 worth of bitcoins a day, and new investors are bringing in approximately $500,000,000 per day, the mining reward is hardly significant in changing the price, because it's such a small percentage of the daily change, probably a lot smaller than the daily variance.
Note that I'm not saying there is 500 million going in and out of bitcoin everyday. I'm using an extreme example to illustrate a point. We don't know how big a % of daily dumps mining rewards are. Because a previous investor exiting the bitcon economy should have the same effect as mining dumps, so what is the ratio of these things?
Again, all else being equal is normally a good way to isolate a variable, but in this case it is not realistic at all. But if it was, bitcoin would still not go up by thousands of percentage points by a few thousand btc a day, at least, it would take many, many years for that to happen. If everything were to stay absolutely constant in those years.