Jun

16

When the numbers look too good, there is an analogy for when one hires a specialist doctor (based on mortality/morbidity stats) or a lawyer (based on courtroom win/loss stats). If a doctor or lawyer has stats that look too good, it is often because he/she doesn't take the toughest cases.  

Ed Stewart writes: 

I wonder to what extent this applies in trading or evaluating traders. Do extraordinary numbers imply something is not what it seems. Certainly the obvious (fraud). but what about situations where it is not that. Do numbers that are too good at times suggest no real money is being made because no risk is present in the program? Reverse engineered to "look good" by metrics but not actually make any money.   

anonymous writes: 

There is a certain quantitative fund led by a renowned mathematician who has supposedly generated persistent returns in excess of 30% for many years.  That fund is not open to outside investors and is (supposedly) available only to employees and partners of the renowned mathematician.  The principals have a number of other funds which are open to outsiders, which have billions under management, and which have produced unremarkable results.

If one were going to set up a clever marketing scheme  one might use this sort of model. One would use the internal fund (with word of mouth only / no audited returns) as the bait.  And then sell the public fund which is vanilla to gather assets.  I am not a lawyer and have no opinion as to the legality.

Another scheme uses the survivor bias:  A manager sets up a series of funds and then closes the worst performing ones. The surviving ones have stellar track records. The manager then markets new funds using the track record of the surviving one. If the funds are segregated, it also produces large amounts of fee income.  A former Salomon Bros forex trader based in Connecticut got in trouble with regulators when he took this to the extreme by opening separately capitalized hedge funds that ran offsetting positions. When one of the funds blew up, the creditors sought to grab assets in the other fund.

A final scheme is what private equity and VC folks always do.  They segregate each series of fund.  They harvest fees from the winning funds but don't give back fees on the losing funds.  Of course if their track is dismal, the game ends. 

John Netto writes: 

Having spent many years living off of my P and L and working closely with quite a few in the Chicago Prop community who have done the same, there are simply strategies which lend themselves to personal wealth generation b/c they have significant capacity constraints and don't scale well. The reality is if you tried to run these at a higher scale it would decay the returns significantly and potentially alter market behavior around those respective trades.  I can say personally that when I'm trading an event with low liquidity getting out a 25 lot on the euro FX futures has a much different dynamic than getting out of a 1,000 lot. A trade which can make 20-30 ticks on the yen can have it's risk-reward profile altered considerably when factoring in liquidity and the velocity of trades around that liquidity.

Also, by exposing the strategy to the public and allowing for the returns to be analyzed you now open the possibility for the Intellectual Property to be compromised through reverse engineering.
So when I hear stories of funds or traders having return profiles like this I'm not surprised at all, even less surprised when they are not available to the public. Analogous to paying $25 for twitter on it's IPO when it traded in the 40s.

Stefan Jovanovich writes: 

What John wrote (thank you!) made me think about its truth regarding war. The big deployments usually produce terrible returns while the small units win the battles.

In the American part of the D-Day landings the mass bombings of the air forces were utterly useless (except to kill French civilians who, to this day, have been remarkably generous about not mentioning the stupidity and honoring the Americans' graves).

The "plan" was to have amphibious-enabled Shermans breach the fortifications. But only half of them even made it ashore; the rest foundered. Of the 66 tanks, 32 made is ashore (27 on Dog, only 5 on Easy). Against those 75 mm barrels the Germans had a roughly equal number of artillery and anti-tank barrels; the problem was that theirs were in reinforced concrete bunkers and pillboxes. Still worse, the artillery was supplemented by 40 rocket-launchers and 85 machine gun nests; against those the men on the beach had only their M-1 Garands.

For an hour and more after landing (H-Hour was 0630) the 1st and 29th Divisions were literally shredded because the Shermans and the combat engineers could not find a way to get them past the fortifications. What saved them was the fact that some individuals followed John's Rules. Even though all naval gunfire support was supposed to end at H-Hour, the 5 destroyers that were part of the Amphibious Assault Group - the Frankfort, McCook, Doyle, Thompson and Carmick - were ordered to close to the beach. (The order could easily have gotten the Destroyers' commander Sanders and the overall Group commander Hall fired for insubordination; under the assault plan all naval gunfire support was to end at H-hour.)

After the battle, James Knight, a Sergeant of the 299th Combat Engineer Battalion, wrote a letter to James Semmes, Captain of the Frankfort: "There is no question, at least in my mind, if you had not come in as close as you did, exposing yourself to God only knows how much, that I would not have survived the night. I truly believe that in the absence of the damage you inflicted on German emplacements, the only way any GI was going to leave Omaha was in a mattress cover or as a prisoner of war." The Chief of Staff of the 1st Division, Colonel S.B. Mason, confirmed as much in the report he wrote after inspecting the German defenses. "I am now firmly convinced that our supporting naval fire got us in; that without that gunfire we positively could not have crossed the beaches."…

Sometimes, good deeds are rewarded. When Hall, the Amphibious Group Commander, retired in 1953 he was still ranked only a Captain, but Eisenhower had him advanced to Admiral "in recognition of his battle honors". To Eisenhower Hall was "the Viking of Assault" (and a fellow football player). Eisenhower undoubtedly knew that, without Hall's, Sanders', Semmes' and the other Navy men's actions, the American part of the landings would have failed.
 


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search