Bitcoin Forum
December 13, 2024, 10:11:35 AM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 [2]  All
  Print  
Author Topic: what can Bitcoin learn from high frequency trading regarding the block size?  (Read 2985 times)
markm
Legendary
*
Offline Offline

Activity: 3052
Merit: 1134



View Profile WWW
September 12, 2013, 01:37:27 AM
 #21

That value can become a self-fulfilling prophecy, too.

The market-making scripts I used to run daily against the Digitalis Open Transactions server basically grabbed such a value, marked it up for sell and down for buy, and posted offers.

If I start running them again, they will pretty much be acting to make manifest the prices put online at http://galaxies.mygamesonline.org/latestrates.inc

-MarkM-

Browser-launched Crossfire client now online (select CrossCiv server for Galactic  Milieu)
Free website hosting with PHP, MySQL etc: http://hosting.knotwork.com/
btcmind
Member
**
Offline Offline

Activity: 98
Merit: 10


View Profile
September 12, 2013, 08:09:20 AM
 #22

HFT just serves the brokers and trading houses. For real investors it is not important that their trade executes within seconds.

Perfect analogy: Bitcoin mining just server the miners. The real users of Bitcoin don't need miners?!

Instead of talking about HFT talk about computer trading and liquidity provision. But for that you will have to understand how a market actually works. I think I will start a thread on that. Good idea, to have actual code to understand what is going. HFT as a term is basically meaningless. HFT simply means you have timestamps. That's all. If you don't have timestamps, you don't know who came first! The alternative is indeed to have batches. But then you would actors predicting cutoffs of batches, which is basically the same. Batches are called call auctions and are commonplace in todays market. You can't have a discussion around this if people don't even have the most basic understand of how stock markets work...

Cryddit
Legendary
*
Offline Offline

Activity: 924
Merit: 1132


View Profile
September 12, 2013, 04:56:04 PM
 #23

Is some truth in idea that high-speed stock trading has some benefits to the instant-auction marketplace.  By playing speed against bid-ask gap they keep bid-ask gap smaller than would otherwise be, so better liquidity, good for if want to buy or sell one or two shares.  But not better than shrinking bid-ask gap to zero.

Is also some truth in idea that high-speed stock trading causes some problems in the instant-auction marketplace.  By playing speed against the bid-ask spread, thins the orders around the gap on both sides.  This force any larger trade to shift market more than otherwise would.  Causes market to shift down faster when someone tries to sell a large order or up faster when someone tries to buy a large order.  Eats into profits of anyone trying to make reasonable-size trades, whether buy or sell, meaning market is LESS liquid for trade larger blocks.  Drives the size of trades always smaller as traders try to avoid making panic or bubble or unduly large loss due to market 'twitch' in presence of flash traders, making it very hard to make a large simple transaction.  Occasionally destabilize market entirely by reacting (wrongly) faster than humans can correct, causing 'flash crash' or 'insta-bubble.'    Increases tendency for the market to respond more to market dynamics than to anything about companies the assets represent.   Finally, tends to cause unfair treatment by brokerages and exchanges; big volume speed trader get good deals smaller traders frozen out of.

Also, is no end to it.  If trades can execute in millisecond, speed trader make more by trading a thousand times a second.  If trades can execute in microsecond, speed trader will make more (tiny tiny percent more) by trading a million times a second.  All problems cause get worse, and only benefit is bid-ask spread shrinking asymptotically toward zero, but that only for orders with size asymptotically approaching zero.  Why bother?  We can reach zero. 

Batch trading improves price efficiency (component of 'liquidity' that traders actually care about) for both small and large trades.  Does not result in twitchiness.  Does not destabilize market.  Does not go out of control faster than human can correct.  P2P marketplace is automatic, no chance of favoritism or corruption.   I believe combination of instant-auction and high speed trading provides no benefit to other traders not better provided in periodic-fair-price model.  Unless buy or sell RIGHT NOW instead of sometime in the next ten minutes is terribly important benefit. 

Finally, and to me most compelling argument; allows humans time to consider situation and respond carefully rather than being in a frantic panic to be faster than neighbor and make stupid mistakes.

Pages: « 1 [2]  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!