If you go by the old British view, a 5% drop in supply will get you a 20% increase in price assuming demand is constant.  If you take a pre-halving price of $450, a delayed halving effect could get you a $1350 price.  The bearish view would be that blocks were already full, so growth is mostly over and you could stop calculations at that point.  On the other hand, you could say this is compensated by the value of each transaction increasing while smaller transactions are pushed off-chain.

http://i.imgur.com/5Oj1bgL.png

Here's where things get dicey.  As you can see above, the number of bitcoins sent per block has been mostly flat for a long time; I mean, if you extrapolate this chart further, it looks kinda the same all the way back to late 2012.  I believe this is explained by a lot of these coin movements being done entirely for things like mixing (to obfuscate sender/receiver).  It's even possible miners attempt to overflow the network they own and run themselves in order to create artificial scarcity of block size and push fees higher.  

Overall, bitcoins sent per block is not that good of a metric since if nobody else is transacting, the miner can send coins virtually for free.  Every block could be full if not even a single person uses the system in other words, and those fabricated blocks could either be dust or large amounts of coins.  

There's also the SHA256 mining exploits that miners use to attempt to save power and mine more efficiently which causes an abundance of empty blocks.  The empty blocks have mostly the same effect as miners purposely flooding blocks with their own free transactions to create artificial scarcity and raise fees higher.  When so many variables and outlier economic strategies are fully in the hands of miners to control, usually at little or no cost to themselves, it really makes many seemingly useful on-chain bitcoin statistics almost worthless.

http://i.imgur.com/jSyKEoq.png

Now we have the dollar value of bitcoin transactions per block.  This looks like a promising statistic that can be interpreted in some type of objective manner at first, but in a time span of only 1.5 months, the value of each bitcoin has doubled.  Clearly not all bitcoin transactions are solely for things like mixing or miners/govt entities attempting to flood blocks to artificially raise transaction fees, but the price increasing so fast would likely double the value of those current actions.  Of course, the value of the coinbase itself would also go from $12.5k to $25k.

So we do know the value of each block has increased, but increasing the bitcoin price also has the side effect of increasing the value of what most would consider spam or non-economic transactions.  Of the transactions that are actually legit commerce, it also doesn't tell you if these people are buyers or sellers.  They're sending mostly the same amount of coins as usual, but are those coins being sent to wallets to hold, or are they being sent somewhere to spend and converted to fiat?

This is where some people will argue with me and claim that since the price is increasing, the supply/demand curve must be shifted way to the demand side.  This is totally false;  price doesn't matter, only liquidity matters.  For example, a coin named NXT at one point had something like a $5-10 million market cap while on the biggest exchange it was traded at (Poloniex), it had only 30 BTC buy support on the entire buy side.  So as you can see, it's possible to create high priced, completely illiquid assets that don't actually have demand.  Conversely, the price of silver has been getting slammed the last several years while demand has actually been increasing.

Long story short, none of these on-chain statistics tell us much of anything.  A better method is to just look at the exchanges themselves.  Bitfinex is currently the market maker exchange of bitcoin.  Anyone who has read my posts before knows I don't trust Bitfinex as far as I can throw them.  What did the start of this price rally look like on Bitfinex?

http://i.imgur.com/96uQox8.png

It's possible bitcoin bucks this trend somehow, but in current form, it kind of appears bitcoin is replicating what's been observed in altcoin markets like NXT before:  jacking up the price while creating illiquid, low volume markets where smaller particpants can still do a little bit of transaction, but larger holders are trapped.  

The price really went parabolic from around the $1200 area, and anyone who has traded this market long enough knows the price usually dumps back down to where it went parabolic from.  In other words, you would likely not want to become a bitcoin whale at this point in time, and this is why you see low volume.  No big money wants to take that possible large downside, questionable upside risk.

Back to my original point.  I said with demand being constant the price of bitcoin could be $1350 post-halving.  But demand wasn't exactly constant; we saw volume fall off a cliff.  Volume was much higher at the $500-800 level cross-exchange.  It further dropped off after the $1000-1200 range.  I believe that old British supply/demand metric might be manifesting itself as the invisible hand of the market through the volume levels, signaling what is a reasonable price with currently capped scalability.

Back when I traded the bitcoin market a few years ago when it was $200, I generally NEVER saw the price move unless there was a giant shorter to squeeze.  The entire trip from $200 to $800 was basically a short squeeze, along with some halving element.  Another interesting bit of information is that the same single entity who controlled the bitcoin price back then on Bitfinex STILL controls the bitcoin price there now.  I watched him use the same exact bot to do certain wall strategies at $400 as right now.  I personally find it horrifying that the price of bitcoin is entirely controlled by one guy/entity on a single exchange for this long, but other people either haven't figured it out or don't care.  

When you saw Huobi leading in price back in the day, it was just fabricated numbers and the market would not move any higher if Finex stopped, so the market really was mostly in the ball court of Bitfinex the whole time.  The rise from $1200 was also a short squeeze from Bitfinex turning off withdrawals and forcing all sellers to become buyers, while all other exchanges kind of followed Bitfinex as a bull run initiator for no logical reason besides possibly believing the cup and handle was in play.

At current price of $2070, I don't know where the upside momentum would come from when no smart traders are really going to buy the top of a parabolic rise, so it feels like you would need yet another short squeeze to go higher.  If people know (I assume I'm not the only person who knows this) that shorters have been getting burned since $200, who the hell wants to be a bitcoin shorter?

(chart 1 - smartbit.com.au, chart 2 - blockchain.info, remaining charts - bitcoinity.com)