UnThreaded | Threaded | Whole Thread (21) | Ignore Thread Prev Thread | Next Thread
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: of 251800  
Subject: MIVLI: MI VL Index Date: 12/9/2012 12:26 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 13
I was going to do a backtest of the top-rated VL screens from the weekly ledger, but then I realized I had no idea how the average VL screen performed.

Back at the beginning of 2006 I introduced a SIPro index; this post looks at a similar index for our VL screens. I went back to the screen pick listings for the end of December 2005 and made a list of screens that also were on Gritton's backtester. SOS and short screens were excluded. Not being familiar enough with VL screens to choose amongst them, I selected every fifth screen. This yielded 27 screens and, with five stocks per screen, an index of 135 stocks.

The screens:
AssRS13
CAPRS
EG5AT
EGPR PE
FOG MI
GAR4CFS
HiDiv
HiYield
KeyEPS
LowPE
LPS1+2 R26
OlapRS52key
PEBsize
PEG NT
PlowPE
PlowEG5 RS631
PST 5/10
REP
RS13wk
RS26wkT12
RSCAP
RSPEG2
SafetyHiccup
Spark
TPEG2
UG
YldEarnYear

Here are the MIVLI results for 2006-11. They are less than inspiring.

MIVLI SPY SIPI
2006 14% 15% 27%
2007 27 5 36
2008 -58 -38 -30
2009 30 29 21
2010 22 15 35
2011 - 7 2 5
.
CAGR 0 2 13

It is not clear that the VL universe adds value, or perhaps we haven't been good at extracting alpha from the VL world. One could argue that the selection of MI screens was too random; perhaps there should be a list of 'recommended' screens.

DB2
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post Back To Top
Author: jakalant Big red star, 1000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240488 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 12:34 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 0
One could argue that the selection of MI screens was too random

More than one would, I would think.

Print the post Back To Top
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240489 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 12:37 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 0
One could argue that the selection of MI screens was too random
---
More than one would, I would think.


How does one create an index?

DB2

Print the post Back To Top
Author: DreadPotato Big gold star, 5000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240490 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 2:32 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 2
How does one create an index?

Why not just use the Value Line ARitemetic or Geometric indices?

VL Arithmetic Index 2006 - 2012 7.1%

OR

You could possible insert all of the screens into the blender or SOS.

I selected about 70-75 and included at least one from a "family". A better choice would probably first run a SOS of each "family" then use that in a blend to reduce many overlaps. Results also marginal for this 2006 - 2011 period.

Blend 3PT_SCV_pst 1-5, AssRS26 1-5, BETA 1-5, BLITZ 1-5, Benchmark 1-5, CAPLOWEG 1-5, CDPD 1-5, DIVIDEND_GROWTH 1-5, EG5PE 1-5, EG5_AT 1-5, EG5_PEG 1-5, EGPLOW_PE 1-5, EGPR_PE 1-5, EG_PELA 1-5, ERS26 1-5, Fundamentals 1-5, GAR4 1-5, GARPEG 1-5, GAR_EG5 1-5, H52EarnPS 1-5, HBSP 1-5, HIGHCASHFLOW 1-5, HIYIELD 1-5, HI_DIV 1-5, HI_INC_CSH 1-5, IN_RS26 1-5, KEYSTONE 1-5, LOWPE_ZLTDA 1-5, LOWPSR 1-5, LPE_YLD 1-5, LowPEsafe 1-5, NoMo 1-5, PEG 1-5, PEG-Minimalist 1-5, PEG13 1-5, PIH4 1-5, PIH_CSO_simple 1-5, PLOW26WK 1-5, PLOWEG5_RS631 1-5, PLOW_PE2MOD 1-5, R13_EG 1-5, REIT 1-5, REV 1-5, ROEPLOW 1-5, RS13WKT12 1-5, RS26WKT12 1-5, RS4WKT12 1-5, RS52WKT12 1-5, RSEG-rgonsal 1-5, RSPEG1 1-5, RSW 1-5, SAFETY_HICCUP 1-5, SLS_RS26 1-5, SPARK 1-5, Screamers 1-5, SomeMoSafe 1-5, TREPPE 1-5, ValueRatio 1-5, YIELD4 1-5, YLDEARNYEAR 1-5, YLDYEAR 1-5, YldDiv 1-5, ZLTDA 1-5

0611BL(3PT_SCV_pst)15(AssRS26)15bt15kb15bi15(CAPLOWEG)15(CDPD)15(DIVIDEND_GROWTH)15(EG5PE)15(EG5_AT)15(EG5_PEG)15qb15(EGPR_PE)15(EG_PELA)15(ERS26)15(Fundamentals)15gr15gp15(GAR_EG5)15(H52EarnPS)15(HBSP)15(HIGHCASHFLOW)15(HIYIELD)15(HI_DIV)15(HI_INC_CSH)15(IN_RS26)15ks15(LOWPE_ZLTDA)15(LOWPSR)15(LPE_YLD)15(LowPEsafe)15(NoMo)15ps15(PEG-Minimalist)15pq15(PIH4)15(PIH_CSO_simple)15bs15(PLOWEG5_RS631)15(PLOW_PE2MOD)15qg15(REIT)15qv15(ROEPLOW)15(RS13WKT12)15(RS26WKT12)15(RS4WKT12)15(RS52WKT12)15qe15po15rw15(SAFETY_HICCUP)15(SLS_RS26)15ss15(Screamers)15(SomeMoSafe)15qt15(ValueRatio)15(YIELD4)15(YLDEARNYEAR)15(YLDYEAR)15(YldDiv)15(ZLTDA)15



Screen S&P

CAGR 3 2

GSD 27 23

Sharpe Ratio 0.17 0.12




gdm

Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post Back To Top
Author: jakalant Big red star, 1000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240491 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 3:06 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
I agree. It is a challenge. I would think that one would have to take into consideration whether a screen is being used. If there is little use it doesn't seem to make sense to include it in an index. That might involve conducting a survey as to which were the top 20(?) VL screens being used. That would probably be of interest to more than a few people here.

Print the post Back To Top
Author: DreadPotato Big gold star, 5000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240493 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 4:23 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 4
Why not just use the Value Line ARitemetic or Geometric indices?

In case you decide to use it this data from VL's database might be useful.

                                                   Total   Total   Total    Total    Total   Total   Total    Total   Total   Total   Total   Total   Total    Total
Ticker Return Return Return Return Return Return Return Return Return Return Return Return Return Return
Company Name Symbol Price 1-Year 3-Year 5-Year 10-Year 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002
CBOE RUSSELL 2000 INDEX RUT 0.00 13.31 (0.22) 8.16 (5.45) 25.31 25.22 (34.80) (2.75) 17.00 3.32 17.00 45.37 (21.58)
Index-Dow Jones Ind Avg DJI 13155.13 10.48 (1.23) 4.55 5.53 11.02 18.82 (33.84) 6.43 16.29 (0.61) 3.15 25.32 (16.76)
NASDAQ 65 COMP. INDEX COMP 2978.04 10.91 13.34 0.81 8.39 (1.80) 16.91 43.89 (40.54) 9.81 9.52 1.37 8.59 50.01 (31.55)
S & P 500 STOCKS SPX 1418.07 10.87 (1.84) 4.78 0.00 12.78 23.45 (38.49) 3.53 13.62 3.00 8.99 26.38 (23.37)
S & P INDEX SMALL CAP 600 SML 15.28 1.23 8.99 (0.16) 24.98 23.78 (31.99) (1.22) 14.07 6.65 21.59 37.53 (15.31)
S & P MIDCAP MID 10.42 14.15 1.55 8.73 (3.10) 24.85 35.00 (37.28) 6.41 9.28 12.10 14.31 34.02 (18.66)
V/L COMP INDEX NEW ARITH WGHT* VLE.I 3075.89 10.78 14.10 4.59 11.89 (5.93) 26.82 60.84 (37.44) 1.30 15.65 6.83 17.23 48.06 (17.11)
V/L INDEX GEOMETRIC COMP* XVG.I 358.26 4.75 7.97 (5.92) 3.29 (11.39) 20.47 36.77 (47.14) (6.64) 10.98 2.00 11.52 37.39 (28.57)
WILSHIRE 5000 INDEX DWCF 14706.78 12.38 11.58 (1.29) 5.75 (0.85) 15.34 25.71 (38.61) 3.78 13.75 4.57 10.67 29.44 (22.08)


gdm

Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post Back To Top
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240494 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 4:50 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
Why not just use the Value Line ARitemetic or Geometric indices?

That would get at the VL part but would miss the MI part: Are the screens adding value? Is your particular selection of screens doing better or worse than the average MI screen?

I would think that one would have to take into consideration whether a screen is being used. If there is little use it doesn't seem to make sense to include it in an index. That might involve conducting a survey as to which were the top 20(?) VL screens being used. That would probably be of interest to more than a few people here.

I'm not sure an index should be a popularity contest. In addition, one would need to know the most popular screens from seven years ago. Those might not remain the popular ones today. (Which brings up a secondary question: are there screens we should tell people not to use? Should they be relegated to a B-list?)

In particular, I was looking at the best and worst screens in the weekly ledger and the question arose, 'did they do better or worse than the average screen?'

DB2

Print the post Back To Top
Author: DreadPotato Big gold star, 5000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240495 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 6:17 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 0
That would get at the VL part but would miss the MI part: Are the screens adding value?

Not sure I understand. How can they be in the top performance ranks and not add value.

Is your particular selection of screens doing better or worse than the average MI screen?

Then again if a screen is out performing the VL index it must be picking good stocks from the VL universe and adding value.

gdm

Print the post Back To Top
Author: aussi One star, 50 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240496 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 7:03 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
For a Value Line index, the index could be the result of holding all Timeliness = 1 for 20 days and averaged over 20 days. Or something similar.

2005 8%
2006 17%
2007 2%
2008 -38%
2009 61%
2010 28%
2011 -5%
2012 15%

URL from GTR1 Blender tim.v:lt7


Craig

Print the post Back To Top
Author: mark19601962 Big red star, 1000 posts Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240497 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/9/2012 7:10 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 0
I think the idea of creating a vl/mi index is great. However, because the average of the screens is not good is not relevant in my opinion, because most people probably only use the better post discovey ones.

That being said, I am surprised the average of all the screens is not better. To be picked as a screen, I thought that you had to beat the averages.

I think one moral of the story is to pick a blend of good post discovery screens that use different strategies. Due to market cycles, even yey could start to underperform.

Print the post Back To Top
Author: gritton Big red star, 1000 posts Top Favorite Fools Feste Award Winner! Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240500 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/10/2012 10:11 AM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 3
perhaps there should be a list of 'recommended' screens.
...
I'm not sure an index should be a popularity contest. In addition, one would need to know the most popular screens from seven years ago.


That's the most difficult point: selecting screens without a hindsight bias. In lieu of "popular", we might assume that the screens that do well are the popular ones. So we could select the top N screens that have done well lately, where "lately" could be a year or a few years. Or we could include all screens that beat the S&P my some margin (10%?). Now the only challenge is trimming the candidate list down to those screens which had been discovered at any given moment. I started to collect such a list a while back, but there were many holes in it.

- Jamie

Print the post Back To Top
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240506 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/10/2012 1:31 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 3
In lieu of "popular", we might assume that the screens that do well are the popular ones. So we could select the top N screens that have done well lately, where "lately" could be a year or a few years.

I've been looking at the best five and worst five VL screens listed in the weekly ledger (five stocks) and, using the results from gritton's backtester, seeing how one would have done with monthly trading. So far I've covered the five years 2003-2007. Here are the results:

CAGR
Best 21%
Worst 23
SPY 12

Both the best and the worst five screens handily beat SPY, although 2008 is yet to come. However, the appears to be no difference between the top and the bottom screens. This would lend some weight to a random selection of screens for an MI index.

DB2

Print the post Back To Top
Author: mungofitch Big gold star, 5000 posts Top Favorite Fools Top Recommended Fools Feste Award Winner! Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240514 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/11/2012 10:57 AM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 6
In the same sense that the SI "index" was used as an example that
quant investing still works and isn't dead, then a VL MI index should be
unbiased to be able to show the same thing. So, no fair using popular screens.
The plurality of practitioners use monthly holds, and we should not make
any assumptions at all about what screens they use.
So my suggestion for a VL index would be an equally weight monthly rebalanced
monthly trading cycle portfolio of all post discovery long VL screens.
About the only variable is how deep; I suggest top 4.
The weekly trade ledgers track 3 and 5, so 4 gives a nice compromise.
That's what Zee used for a lot of his "what screen" research.
A lot of research shows that a lot of the performance is concentrated
in the topmost ranks, so I prefer 4 to 5.
"Post discovery" is easy if it's generated from the weekly picks.
Since the purpose is to see what the typical screen would have done, you'd of
course multiply-weight things picked on multiple screens—a blend, not an SOS.

In an ideal world I'd make the index level equal to the average dollar portfolio
value of four 4-week-hold portfolios trading at market close on the
first trading day of each week.

Jim

Print the post Back To Top
Author: FlyingCircus Big red star, 1000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240518 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/11/2012 7:30 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
My opinion - we can't draw conclusions from a sample of screens people here developed 7 or more years ago with heavy momentum focus / earnings focus and misdirect it as "The VL universe isn't adding value."

Suggest going back to Zeelotes' studies from 2007-2009 of the Gold screens. The Gold screens could be the index (there were @30?). Or, as every other index changes, the best screens rankings by Sharpe/GSD, Sortino, that he posted back early 2009 and early 2010.

This is, obviously, a little bit of tedium - but it's better than randomly selecting from our VL screen universe, which is littered with the debris of over-mined, time-specific ideas that have not held up out of sample.

FC

Print the post Back To Top
Author: mungofitch Big gold star, 5000 posts Top Favorite Fools Top Recommended Fools Feste Award Winner! Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240523 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/12/2012 4:47 AM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 5
This is, obviously, a little bit of tedium - but it's better than
randomly selecting from our VL screen universe, which is littered with
the debris of over-mined, time-specific ideas that have not held up out of sample.


I'd conclude the exact opposite from "better".
If you cherry pick the things to put into your so-called index you learn nothing,
it's just another tuned blend. If everything goes in, we learn whether
or not the collection of VL long screens as a whole have added value post discovery.
We might also learn things about when VL long screens are doing well and when they aren't.

Perhaps we're scared of doing that because we already know the answer.
When DrBob posted about the SIpro "index", no one suggested it was
invalid or inappropriate because it included all of the screens.
That's because the numbers looked good.

Jim

Print the post Back To Top
Author: Zeelotes Big red star, 1000 posts Feste Award Nominee! Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240547 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/14/2012 2:26 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 6
Dr. Bob wrote:
That would get at the VL part but would miss the MI part: Are the screens adding value? Is your particular selection of screens doing better or worse than the average MI screen?

My personal take on this question of an MI Value Line Index is that no one in his right mind would invest in an index containing all of the MI screens developed and shared here.

In a sense, everyone cherry picks what it is that they are going to use to invest in. Hopefully, that cherry picking, as Jim has pointed out in this thread is:

http://boards.fool.com/blending-at-a-whole-new-level-updated...

If you're going to pick stocks mechanically and use backtests to
determine what methods work well, it makes sense to pick your screens
mechanically and use backtests to determine which of those work well.


It is foolishness in my view to judge the effectiveness of the work that has been done here by comparing it to an index of all screens tracked. Part of the work done on this board has been related to the wisdom of using blends. In the early years attempts were made to create a blend and they were called a Screen of Screens (SOS). Later this morphed into using the Excel Solver. Hopefully, my posts back in 2007 raised this to a whole new level, where a backtest of screen selection guided the selection of the screens to use.

Here is Dr. Bob's table including the returns from a blend:

 MI   VLI  SPY  SIPI  Blend
2007 27 5 36 6.64%
2008 -58 -38 -30 1.84%
2009 30 29 21 61.40%
2010 22 15 35 25.00%
2011 -7 2 5 7.93%

CAGR 0 2 13 21.10%

NOTE: The 2007 return for the blend is from the PEAK in 2007 so it is not an apples to apples comparison.

Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post Back To Top
Author: FlyingCircus Big red star, 1000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240553 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/14/2012 9:44 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 0
If you cherry pick the things to put into your so-called index you learn nothing,
it's just another tuned blend.


So, as usual, your point is valid. Maybe the most important lesson learned is that most of the screens were over-tuned for the conditions under which they had been recently successful?

If the "index" returns from a random sample are generally mediocre, then does it not follow that most of the screen definitions and development were overtuned, mediocre and/or meaningless?

Which would lead to the question, why are we still carrying around a pile of overtuned, mediocre and/or meaningless/torpedoed screen definitions?

FC

Print the post Back To Top
Author: Rayvt Big gold star, 5000 posts Top Favorite Fools Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240555 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/14/2012 11:44 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
Which would lead to the question, why are we still carrying around a pile of overtuned, mediocre and/or meaningless/torpedoed screen definitions?

Amen to that!!
I suspect it's because the people who report the screens picks have it automated, so it's no effort to keep them all, but it would be an effort to remove some screens.

But, yeah, every time I see that post with lists of screens that scroll down and down before you get to the bottom ... I just shake my head.

Print the post Back To Top
Author: mungofitch Big gold star, 5000 posts Top Favorite Fools Top Recommended Fools Feste Award Winner! Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 240564 of 251800
Subject: Re: MIVLI: MI VL Index Date: 12/15/2012 12:52 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 19
If the "index" returns from a random sample are generally mediocre,
then does it not follow that most of the screen definitions and
development were overtuned, mediocre and/or meaningless?
Which would lead to the question, why are we still carrying around a pile
of overtuned, mediocre and/or meaningless/torpedoed screen definitions?


For the exact reason that you state:
So we know that most of the screens are overtuned dross.

It allows the possibility of saying what the "average" MI screen has
done post discovery, which we could not do if the list were pruned.
And it's an ideal inoculation against hubris or false optimism on screens, new or not.
It's not like carrying around the bad ones is a lot of work for anybody;
the process is a bit of work, but no more work for 100 screens than 50.
Thus, I strongly believe the list should not be pruned.

However, I wouldn't object to a fresh consensus ranking them all gold/silver/bronze/iron/arsenic,
so newcomers can tell which end of the farm tends to have the best crops.

Jim

Print the post Back To Top
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 241219 of 251800
Subject: Re: MIVLI: MI VL Index Date: 1/24/2013 1:05 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 1
Jamie's site has the performance number for 2012. The MIVLI turned in a 12% performance for the year; the S&P500 over the same dates was 14%.

DB2

Print the post Back To Top
Author: DrBob2 Big funky green star, 20000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: 247989 of 251800
Subject: Re: MIVLI: MI VL Index Date: 1/20/2014 9:31 AM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 2
Jamie's site has the performance number for 2013. The MIVLI turned in a 45% performance for the year; the S&P500 over the same dates was 32%.

DB2

Print the post Back To Top
UnThreaded | Threaded | Whole Thread (21) | Ignore Thread Prev Thread | Next Thread
Advertisement