Wednesday, April 10, 2019

The currency banker and unexpected disasters

There are things called known unknowns, like the variance between supply and demand.  We can say, with certainty, that supply varies with demand typically 3%, for example, over a seasonally adjusted range. That known unknow is what is discovered, in the pits. It is not the pit bosses business to gain or lose in that game, it is neutral. The pit boss keeps the variance at 3%, letting it bounce to 5% for short periods rarely; and the pit boss does this by algorithm and contract.

Unexpected disasters are shared by all and appear in the currency risk.  The currency banker will gain on a natural disaster since money is forced to take round about paths during the chaos. That wandering money is covered in the deposit balance by the pit boss, and it gains currency from the market.

It is always a two step process. An unexpected loss in productivity results in gain to the pit boss currency account which then discharges that gain to recover bounds.  The reverse for an unexpected productivity gain.  But in round robin fashion, the first finder is not always the pit boss.  The unexpected is generally covered before the pit boss gets a go since the board is equally observeable to all.  A shift in the currency account results in adiabatic recovery by the pit boss, and in those cases the market will pass.

It is a type of trading game.  A bot observing the board discovers a large discrepancy.  If the discrepancy it too large then hedging that discrepancy is out of trading bounds, the bot will demur. The reason is simple, it knows the pit boss will run almost next, and make an adjustment.  The mis-sampled result is too chaotic to bet. The bot may make a bid and be interrupted by the pit boss, according to contract, then the bot is frigged.

Read he fine print on the pit contract.  It states:
"Upon one trading bot finishing its pass through the board, the executive shall first check if discrepancy is out of bounds and run the pit boss if so"

That clause can be risky if the entering bot had prior assumption of the prior state of the boards, what was the variance going in and what is it now.  If there was a large change, then the bot paid a wasted board fee to check a possible hedge, and the pit boss must have run and repaired it.

Like bees swarming, traders suddenly find themselves congested because of an obvious hedge, but the contract puts the pit boss first in line for large discrepancies.  In an optimally congested system, the number of save paths is a finite set with few hedges, because the queues are short; combinatorials few.  The trading bot always prefers inside information, and that is what currency bankers monetize (or demonetize).


No comments: