Use access key #2 to skip to page content.

portefeuille (99.78)

20 minutes delay cancel feature

Recs

8

May 27, 2010 – Comments (22)

That's it. I am really getting tired of geeting enraged by this bug (see comments #20,57 here).

I will not make another call in the "caps" game with any of my 12 (hehe) players until this bug is fixed. Feel free to join me. Or recommend this post. Or at least think about it for a second and get angry as well. Then do whatever you feel like doing.

 

22 Comments – Post Your Own

#1) On May 27, 2010 at 6:04 PM, Starfirenv (< 20) wrote:

One of the reasons I choose not to "play". +1 Rec

Report this comment
#2) On May 27, 2010 at 6:21 PM, portefeuille (99.78) wrote:

#20,57

#29,57

 

--------------------

The ability to "cancel picks during 20 minute delay" can be misused quite easily.
Consider the following:

create a new player.

use the following "script":

n = 200
WHILE n>0 DO
BEGIN
give "outperform" ratings to the n stocks/ETFs ratable in "caps" that display the highest short term volatility at that particular moment.
get real-time quotes for these n stocks/ETFs and the s&p.
wait for 19 minutes (just to make sure ...).
get real-time quotes again.
calculate o_i := outperformance of stock/ETF i.
FOR i = 1 to n DO
BEGIN
  IF o_i < THRESHOLD1 DO
  BEGIN
    cancel the pick for stock/ETF i.
    counter := counter + 1
  END
END
n := counter
counter := 0
END

For a THRESHOLD1 of say 5% it should not take more than a couple of hours for the portfolio to fill with 200 picks giving the new player a score in the area of 1000-2000 points and an "accuracy" in the 90s (depending on the volatility and the "noise")
After 7 days start closing the picks the "opposite" way, now using a THRESHOLD2 of say 3% (the player's picks are no longer "highest in volatility", so THRESHOLD2 should be chosen somewhat smaller than THRESHOLD1).
For every closed pick select a new one (do the abovementioned trick for n=1).
The actual performance of the stocks/ETFs is of no importance (it is the background "noise", the score comes from buying at a discount  greater than THRESHOLD1 and selling at a discount greater than THRESHOLD2.

The new player should rise to the #1 spot in the "caps" game within 5-10 weeks depending on the market volatility (the higher the better).
Creating more than one new player and some optimisation of the "script" would certainly speed things up.

--------------------

(from comment #10 here)

 

So if you have nothing better to do and like writing little scripts, please demonstrate that it works.

 

Report this comment
#3) On May 27, 2010 at 6:21 PM, USNHR (35.92) wrote:

The 20 minute delay is so that you don't try to play the swings of the daily market. It is supposed to encourage a long term attitude. I don't have a big issue with it. It might cost me a couple of points here and there, but overall I don't worry about it to much.

Report this comment
#4) On May 27, 2010 at 6:22 PM, Option1307 (30.45) wrote:

I agree that the feature does have problems, but one thing at a time Port. Let's jsut be excited that they are "fixing" the other problem first.

Report this comment
#5) On May 27, 2010 at 6:23 PM, Option1307 (30.45) wrote:

I'd also love to be able to edit blogs, at least for 5-10 minutes...

*jsut --> just

Report this comment
#6) On May 27, 2010 at 6:23 PM, USNHR (35.92) wrote:

Oh yeah, and half the time when I'm copying your picks, I don't know what they are until the next day, so I have a 24 hour and 20 minute delay..

 Ha ha ha....

Report this comment
#7) On May 27, 2010 at 6:27 PM, portefeuille (99.78) wrote:

#3,4,6 The bug is the ability to cancel the starting /ending of calls within the first (usually around 20) minutes.

#5 I agree.

Report this comment
#8) On May 27, 2010 at 6:27 PM, Option1307 (30.45) wrote:

The 20 minute delay is so that you don't try to play the swings of the daily market. It is supposed to encourage a long term attitude.

I understand the rational behind it and agree with it, but I think the real issue is that you can cancel a stock within the 20 min. wait period if it starts swinging in the "wrong" direction. I'd be fine with keeping the 20 min. wait period for opening/closing, but maybe just eliminate the ability to cancel these "orders".

Report this comment
#9) On May 27, 2010 at 6:30 PM, portefeuille (99.78) wrote:

I think the pseudo code algorithm in comment #2 above should make it pretty clear that there is no way this can be left as it is. Unless of course you assume that the players do not use it, but looking at what the players are doing to "pimp" their results that would be just an illusion, I guess ...

Report this comment
#10) On May 27, 2010 at 6:31 PM, portefeuille (99.78) wrote:

#7 = #8 exactly.

Report this comment
#11) On May 27, 2010 at 6:40 PM, portefeuille (99.78) wrote:

Letting people make their bets after the event is simply never a good idea.

The Sting

Help! I can't find the CANCEL button on my brokerage account!

Report this comment
#12) On May 27, 2010 at 7:08 PM, TSIF (99.96) wrote:

The 20 minutes is not a hard, fast rule.  It can vary.  Those playing with it are just as likely to get burned on a fast moving stock and lose a few percent.  It's a useful tool if you do wrong way thumb one or pick the wrong ticker. Kind of like your brokerage account asking if you are sure and putting up several screens to show what you are doing before you committ. Anyone playing it for CAPs points is pretty desparate. I seriously doubt if your computer script would benefit them.  So they have a one time better entry point than they might have, and might squeeze a little juice out on a close, but if volitility is high, I doubt they can get out of the pick with the 7 day rule in any better shape than the average player. They would certainly have a lot of JUNK in their portfolio, and certainly couldn't close many of the picks 7 days later in the positive.  I know you hate the accuracy rule as well, but this scripted player would be crushed by it.  More than saturating his 200 picks with RED calls, reducing slots for new calls, etc.

In the days before CAPS "Limit Orders" I think the 20 minute rule  had it's advantages.  In the real world you could set a limit order, but you couldn't on CAPS.  You can set a limit order in real life and change it before it activates. You can set a limit order on CAPS and change it before it activates.  You an do trailing stops, etc, in real life.

Also, when the rule was started, having real time quotes was rare and cost a premium.  I think the purpose there was to even the playing field between the haves and have nots. CAPS didn't have real time quotes from their feeds which cost a huge premium.

That said, while it seem rather benign to me, for those purists, I do agree it's probably outlived it's usefulness...assuming it ever had one, which is just my opinion and perhaps faulty memory.

 It does appear that with Limit Orders tested, approved and in use by those who want to use them AND with Real Time quotes more the norm than the exception, that the 20 minute delay could be done away with.  Limit orders are my tool now to ensure I don't wrong way thumb an equity and I can check them for accuracy while I patiently wait to see if they get invoked.  (Using the term patiently as a goal, not reality!)

#4, Option1307, shares my viewpoint. One problem at a time, the more serious ones first. The volume fix should help even the game if some other quirks don't result.

PS Porte, Did you know that if you misspelled words and put them on a flash card, that 90% of the time or higher, the reader doesn't even notice.   Of course if you hold up a correction card, the reader does have to process what the message is/was and mental throughput is drastically cut!!!  :)

 

 

Report this comment
#13) On May 27, 2010 at 7:20 PM, portefeuille (99.78) wrote:

The 20 minutes is not a hard, fast rule.  It can vary.  Those playing with it are just as likely to get burned on a fast moving stock and lose a few percent.

No, that would be no problem. If you notice that the cancelation period has ended for one of the "pending" calls you can still cancel all other calls and the one call that got "executed" was randomly chosen, so the result of these cases is "net neutral".

Kind of like your brokerage account asking if you are sure and putting up several screens to show what you are doing before you committ.

Then that is what they could do instead of letting this backdoor stay open.

Report this comment
#14) On May 27, 2010 at 7:22 PM, portefeuille (99.78) wrote:

No, that would be no problem.

I am talking about the algorithm there. It still works.

Report this comment
#15) On May 27, 2010 at 7:24 PM, TSIF (99.96) wrote:

How is this backdoor any different than limit orders?  Guess cause I live in the country and always leave my backdoor unlocked, unless the bears are around, I don't see the problem.  The 7 day rule and the accuracy hits keep volume trading down.  Nudging a few percent here or there doesn't pay for the effort. Anyone doing this without a scripted program will get burnt, or burned out FAST!

Report this comment
#16) On May 27, 2010 at 7:27 PM, BuffetsMentor (< 20) wrote:

god damn, i didnt know people put this much time and effort into caps picks

Report this comment
#17) On May 27, 2010 at 8:02 PM, portefeuille (99.78) wrote:

If the "20 minute delay cancel period" were shorter than 19 minutes frequently (I do not think it is) you would need to optimise the algorithm a little by for example "cancelling the worst calls early". Once you have found out how the length of the cancel period is distributed it should be quite easy to come up with a "satisfactory" solution.

Report this comment
#18) On May 27, 2010 at 8:08 PM, portefeuille (99.78) wrote:

They would certainly have a lot of JUNK in their portfolio, and certainly couldn't close many of the picks 7 days later in the positive.

A bunch of high volatiliy stocks is not a bunch of "JUNK". That would be a little easy ...

To get rid of any bias that might arise from the choice of "high volatility stocks" you could make about 50% "underpeform" calls (with appropriate changes to the algorithm).

Report this comment
#19) On May 27, 2010 at 8:29 PM, portefeuille (99.78) wrote:

(to ignorant "staff") idiots.

okay, I feel better now. leave this bug in place. I will still refrain from making new calls. a good decision even if they did change something.

Report this comment
#20) On May 27, 2010 at 8:31 PM, portefeuille (99.78) wrote:

I should spend the gained time on studying grammar and spelling of the English language.

Report this comment
#21) On May 27, 2010 at 9:50 PM, ChrisGraley (31.54) wrote:

You think that's a bug! How about when you pick a low volume stock and the pick stays open for a week but the price changes during that time. If the price changed then why wasn't the pick executed? There had to be some volume for the price to change!

Report this comment
#22) On May 27, 2010 at 10:12 PM, Option1307 (30.45) wrote:

There had to be some volume for the price to change!

Ya that happens to me too sometimes. I used to think it was b/c the stock would drop below price/market cap limits during the movement, but that awsn't the case. I guess Caps just doesn't like low volume. Discrimination I say! ha, jk.

Report this comment

Featured Broker Partners


Advertisement