Current HYP top 10 or 20 pick-list

[DiamondEcho]

Can someone who knows what the various selection metrics are have a go at posting the subject?

I'm going to have some funds coming in shortly, from the buy-out of an existing holding that need to find a new home, and if possible would prefer not to have to reinvent the wheel.

[Gengulphus]

Can someone who knows what the various selection metrics are have a go at posting the subject?

Sorry, but there's no such thing as "the various selection metrics", just general guidelines about the shares:

* having high yields - but just which form of yield (historical, forecast, etc) and how big it has to be to count as "high" is at the discretion of the HYPer;

* having properties that make the HYPer believe its dividend is sustainable (and preferably growable) - but what makes the HYPer believe that is again at the discretion of the HYPer;

* being well-diversified - and you've guessed it: the details of that (such as how many companies/sectors you want, and when you should count companies as being in the same sector) are also at the discretion of the HYPer.

And only you know what your various selection metrics are and so are able to select the current top 10 or 20 according to them... Or maybe you don't, in which case you're putting the cart before the horse: you need to decide what they are before doing the selection.

Of course, a pedantic interpretation of that says that you can use your discretion to regard any method at all of selecting shares as "HYP". And indeed you can - but once you post to say what you regard as "HYP", you expose yourself to disagreement and criticism (which are both inevitable and IMHO an essential part of having a proper discussion), or if the way you use it is too unreasonable, even derision or moderation...

So for instance, whether someone thinks that a yield should exceed the FTSE100 yield, the FTSE350 yield, the FTSE AllShare yield, the Equally-weighted FTSE100 yield or doubtless various other reasonable possibilities to count as "high" is a matter for discussion, and one risks some such discussion no matter what one says about it - I've been taken to task by some for describing a share's yield of around 3.3% as not high enough, on the grounds that it was comfortably above the Equally-weighted FTSE100 yield, but I'm sure that I would have been taken to task by others if I had instead described it as high enough, on the grounds that it was well below the FTSE100 yield... And both would be reasonable discussion for the high-yield boards, that might help a reader decide how they felt about what counts as a high yield - but if someone were to try to maintain that a yield of 0.1% was "high", their post would be liable to be moved to some other board!

Gengulphus

[TUK020]

DiamondEcho,
What I found really, really helpful, was a 'worked example' to select a portfolio of HYP shares that was published on the old site by none other than Gengulplus in Dec 15.
I cut and pasted the content into a note pad to keep handily. Perhaps G+ can provide a link to the article.
I do hope that he will update the analysis sometime soon, but appreciate that it looks like a lot of work and thought.
tuk020

[Itsallaguess]

What I found really, really helpful, was a 'worked example' to select a portfolio of HYP shares that was published on the old site by none other than Gengulplus in Dec 15.

I cut and pasted the content into a note pad to keep handily. Perhaps G+ can provide a link to the article.

JohnnyCyclops also carried out a few HYP Screen processes publicly back on TMF, and I know a few were caught for inclusion in the waybackmachine before TMF went down.

Here's a link to one of them, so it'll be worth a read for anyone interested.

https://web.archive.org/web/20161115160 ... sort=whole

With regards to the worked example from Dec 15, if that wasn't caught in the waybackmachine then I don't think there will be any harm in reproducing the contents of your notepad file if you've got a copy of it, although it might be worth Gengulphus confirming if he has a waybackmachine link to it initially?

Cheers,

Itsallaguess

[Breelander]

What I found really, really helpful, was a 'worked example' to select a portfolio of HYP shares that was published on the old site by none other than Gengulplus in Dec 15...

Another 'worked example' was published in 2007 by 'The Investor' on Monevator. A number of Monevator writers also wrote for TMF, so they were very familiar with the HYP concept.

Previous posts in this high yield share investing series have argued the case for dividends, considered what makes a good high yield share, and stressed the need for diversification. Now it’s time to make yourself a cup of tea and settle down to see exactly how you can construct a high yield portfolio for yourself.

http://monevator.com/selecting-the-shar ... yp-part-4/

[DiamondEcho]

JohnnyCyclops also carried out a few HYP Screen processes publicly back on TMF, and I know a few were caught for inclusion in the waybackmachine before TMF went down. Here's a link to one of them, so it'll be worth a read for anyone interested. ...Itsallaguess

Thanks for that, very useful as a refresher on the metrics. And the lay-out has a familiar ring to it, it would either be this post and/or one from TJH from which I extracted the outline 'sift-metrics' that I've had posted at the top of my HYPTUS/S for years. As a summary note for me to use as a base to work from; which means not 'reinventing the wheel' every time I look for a new pick:

Post dated: October 2013 [for comparison, and if I wish to try and update to the present].
'Market Capital: above £1bn
Dividend Yield above 3.2% (current rough FTSE100 yield)
P/E: below 17.0 (current rough FTSE100 P/E)
Dividend cover: above 1.5
Four year dividend growth average above 5% (DL's maximum period is 4 years, I'd have preferred 5).

JC then created a table under these headings:
'Name / Sector / Market Cap (m) / Div. Cover / Avg Div. Grth[4yr] / Div Yield / P/E

Sorting below in alphabetical order by the share name he notes:
'Here's the table of the 34 stocks that got through the screen parameters, sorted alphabetically. I've added sectors, which shows a good mix, many of the HYP 'usual suspects'.
Writing this post and formatting the table took longer than running the HYP screen on DL. For those looking for a 'light touch' simple approach to HYP investing then using screens as a starting point is the way to go. There are probably a few more checks to make before buying anything (perhaps debt levels, and checking if the dividend growth percentage masks recent cuts), but honestly, buying a 15/20/25 holding/stock mix of the following isn't going to be the most dangerous thing one can do this year!'

He then goes on to distil a focused-15 list, with Div Yield% and P/E:>
'Sorting by yield (high to low) and picking the first from each sector, gets the following 15 picks. Personally, I'd not choose Ladbrokes (because I know it cut its dividend in 2009 but has grown it strongly since) or Imperial Tobacco (personal choice). As I noted above, this would need further research before deciding whether to buy.'

Interestingly JC said he had the above set up in a 'screener' within Digitallook. I use the latter but haven't noticed a screener there before - time to investigate! [edit on this point: I've found 'Screener alerts on DL, but they are for subscribers only. Hence I'll have a go and sifting manually].

I wonder how such a focused-15 list would look today, sifted per the same metrics. Well we'll see If it comes out as like a Focus-3 list or similar perhaps that will reveal what if any metrics need to be updated.

[Raptor]

Whoops deleted first post as forgot one of the criteria.

You mean something like this:-



No checking done, just entered the criteria.

Raptor.

[monabri]

PFG...hmmmm! I think we can safely remove PFG from the list.

[Raptor]

PFG...hmmmm! I think the site needs an update.

It says data last updated 1st November 2017 and I used Telegraph prices as not sure the Yahoo one would work... As always DYOR.....

Not sure why SKY got through either......

Raptor.

[monabri]

Sorry Raptor, I did edit my post but you got in there first before I re submitted. Either way, Provident retracted the interim divi and the final is very unlikely. So, as you say, DYOR.

[DiamondEcho]

For my sift I started with the FTSE-100, plus the top-50 by MktCapn from the FTSE-250
Sorted by Div Yld% [aware this likely misses Special Divs - a point to revisit?]
Deleted sell, neutral and underweight broker rated.
IDK the current FTSE-100 average yield%[FY%], but deleted anything under 4.5%
Then took out RDSA as it runs parallel to RDSB which most UK investors prefer.
[Trying to format tables never works for me on TLF, but here is the raw top-15 sift, sorted by DivYield%, down.]
Note: F1 = FTSE100, F2 = FTSE250.
---------------------------------------------------------------------------------------------
Index Name Div Yield P/E Market Cap (m) Brk-rec
F1 SSE 6.71% 10.8 £13,894.25 Buy
F2 Phoenix Group Holdings (DI) 6.63% -22.2 £3,000.31 Overweight
F1 BT Group 6.22% 8.6 £24,571.29 Overweight
F1 GlaxoSmithKline 6.07% 12.9 £64,828.23 Overweight
F1 Vodafone Group 5.99% 30.6 £58,166.11 Buy
F1 BP 5.98% 1,096.1 £100,405.99 Overweight
F1 Royal Dutch Shell 'B' 5.80% 55.9 £92,045.34 Overweight
F1 Legal & General Group 5.41% 11.9 £15,795.16 Overweight
F1 Imperial Brands 5.37% 11.9 £30,384.94 Strong Buy
F1 British Land Company 4.92% 15.6 £5,977.40 Overweight
F2 HICL Infrastructure Company Ltd 4.88% 12.6 £2,804.04 Strong Buy
F1 Aviva 4.70% 32.4 £19,883.55 Buy
F1 ITV 4.59% 9.2 £6,276.59 Overweight
F1 Hammerson 4.58% 17.9 £4,156.50 Overweight
F2 Investec 4.56% 10.4 £3,372.29 Overweight

* This hasn't yet gone through the '4/5 years div growth hurdle' yet.

[Hypster]

I've reformatted it for you DiamondEcho.


Note: F1 = FTSE100, F2 = FTSE250.

[Raptor]

Note: F1 = FTSE100, F2 = FTSE250.

What about:-

P/E: below 17.0 (current rough FTSE100 P/E)

See above table.....

Raptor.

[Hypster]

Out of interest I've attempted to re-create JohnnyCyclops's screen in SharePad:

1. Market Capital: above £1bn
2. Dividend Yield above 3.2%
3. P/E: below 17.0
4. Dividend cover: above 1.5
5. Four Five year dividend growth average above 5% (DL's maximum period is 4 years, I'd have preferred 5). I've used 5*.

* I assumed he meant annualised dividend growth rate of at least 5% in each of the last five years for his last filter.

[DiamondEcho]

Sorry, technically-challenged, really appreciate the help to correct/sift it down!

[Gengulphus]

What I found really, really helpful, was a 'worked example' to select a portfolio of HYP shares that was published on the old site by none other than Gengulplus in Dec 15.
I cut and pasted the content into a note pad to keep handily. Perhaps G+ can provide a link to the article.

I can supply links to the archived copies of the threads for four such selection processes that I did - it's presumably the last that you're referring to, but I might as well supply them all:

"Lump sum all at once" HYP (August 2011): https://web.archive.org/web/20161114200 ... sort=whole

UHYP13 (November 2013): https://web.archive.org/web/20161115210 ... sort=whole and https://web.archive.org/web/20161217110 ... sort=whole

UHYP14 (November 2014): https://web.archive.org/web/20161217130 ... sort=whole and https://web.archive.org/web/20161217131 ... sort=whole

UHYP15 (December 2015): https://web.archive.org/web/20161213103 ... sort=whole and https://web.archive.org/web/20161217143 ... sort=whole

The first was always meant as a one-off, and I've never seriously revisited it since, so don't know exactly how well it's done since. Looking at its list of shares just now, though, I see that it contains Carillion, Centrica, Mitie, Pearson and Tesco, so I wouldn't get my hopes too high! If anyone else wants to try, none of its companies has been taken over since it was selected, which means you won't encounter the major 'hindsight' problem created by trying to select takeover replacements long after the fact, and do note that you should use the revised selection that includes Imperial Tobacco/Brands that appears later in the thread - no tobacco company appeared in the initial selection as the result of a data error that only came to light later.

The last three were intended as the start of a roughly-yearly series, and I did buy them in "1/8th scale model" form and review UHYP13 and UHYP14 when later ones in the series were selected - links are https://web.archive.org/web/20161217132 ... sort=whole (UHYP13 reviewed in 2014) and https://web.archive.org/web/20161217143 ... sort=whole (UHYP13 and UHYP14 reviewed in 2015). UHYP13 was doing rather badly at that time and UHYP14 rather averagely, and a quick look at their holding lists doesn't suggest any dramatic improvements are likely.

Unfortunately, TMF closed their boards about this time last year, and while TLF was set up fantastically quickly, it was some time before tables and polls were working and stable - these were essential features for the UHPxx selection process. And it was longer before Christmas/New Year and their distractions were past and I'd learnt to use tables effectively enough to avoid nasties like columns running into each other in tables with a lot of them. Furthermore, the amount of time/energy I had available for concentrated work was somewhat limited, and there was a major competitor for it in the form of locating and archiving TMF threads that I didn't want to lose. Also, I have to admit to dithering a bit about whether to use the improved TLF polling facilities (N out of up to 10, rather than 1 out of up to 5) to improve the polling part of the process, and if so, exactly how. The net result is that selecting 'UHYP16' got delayed long past the point where it could reasonably be called that...

And although I could in principle skip 'UHYP16' and just move on to selecting 'UHYP17', I'm again in the position of having too much other stuff I want or need to do more, partly as a result of a recent medical problem. So I'm afraid it's time for me to admit that I bit off more than I could chew with the idea of the UHYPxx yearly series, and abandon it. At least hopefully that will make it easier for me to catch up on the backlogs I've started accumulating on GDHYP and CHYP1...

Gengulphus

[CryptoPlankton]

Out of interest I've attempted to re-create JohnnyCyclops's screen in SharePad:

1. Market Capital: above £1bn
2. Dividend Yield above 3.2%
3. P/E: below 17.0
4. Dividend cover: above 1.5
5. Four Five year dividend growth average above 5% (DL's maximum period is 4 years, I'd have preferred 5). I've used 5*.

* I assumed he meant annualised dividend growth rate of at least 5% in each of the last five years for his last filter.

I think there may be a problem with that screening as Centrica fails to meet condition (5)?

[Gengulphus]

Trying to format tables never works for me on TLF, but here is the raw top-15 sift, sorted by DivYield%, down.

Something you might try is pressing the " button in Hypster's post above and studying what appears in the input window - it contains what he used to create his table (between the table and /table tags), and it's nicely formatted to be human-readable (unlike some of the spreadsheet-generated tables I've been known to produce!). When you're done studying it, of course, click on something other than "Submit" to avoid posting a pointless quote of his post with nothing extra added.

Sometimes a clean real-life example of something being done is worth any amount of technical explanation about how to do it!

Gengulphus

[TUK020]

What I found really, really helpful, was a 'worked example' to select a portfolio of HYP shares that was published on the old site by none other than Gengulplus in Dec 15.
I cut and pasted the content into a note pad to keep handily. Perhaps G+ can provide a link to the article.

I can supply links to the archived copies of the threads for four such selection processes that I did - it's presumably the last that you're referring to, but I might as well supply them all:

"Lump sum all at once" HYP (August 2011): https://web.archive.org/web/20161114200 ... sort=whole

UHYP13 (November 2013): https://web.archive.org/web/20161115210 ... sort=whole and https://web.archive.org/web/20161217110 ... sort=whole

UHYP14 (November 2014): https://web.archive.org/web/20161217130 ... sort=whole and https://web.archive.org/web/20161217131 ... sort=whole

UHYP15 (December 2015): https://web.archive.org/web/20161213103 ... sort=whole and https://web.archive.org/web/20161217143 ... sort=whole

The first was always meant as a one-off, and I've never seriously revisited it since, so don't know exactly how well it's done since. Looking at its list of shares just now, though, I see that it contains Carillion, Centrica, Mitie, Pearson and Tesco, so I wouldn't get my hopes too high! If anyone else wants to try, none of its companies has been taken over since it was selected, which means you won't encounter the major 'hindsight' problem created by trying to select takeover replacements long after the fact, and do note that you should use the revised selection that includes Imperial Tobacco/Brands that appears later in the thread - no tobacco company appeared in the initial selection as the result of a data error that only came to light later.

The last three were intended as the start of a roughly-yearly series, and I did buy them in "1/8th scale model" form and review UHYP13 and UHYP14 when later ones in the series were selected - links are https://web.archive.org/web/20161217132 ... sort=whole (UHYP13 reviewed in 2014) and https://web.archive.org/web/20161217143 ... sort=whole (UHYP13 and UHYP14 reviewed in 2015). UHYP13 was doing rather badly at that time and UHYP14 rather averagely, and a quick look at their holding lists doesn't suggest any dramatic improvements are likely.

Unfortunately, TMF closed their boards about this time last year, and while TLF was set up fantastically quickly, it was some time before tables and polls were working and stable - these were essential features for the UHPxx selection process. And it was longer before Christmas/New Year and their distractions were past and I'd learnt to use tables effectively enough to avoid nasties like columns running into each other in tables with a lot of them. Furthermore, the amount of time/energy I had available for concentrated work was somewhat limited, and there was a major competitor for it in the form of locating and archiving TMF threads that I didn't want to lose. Also, I have to admit to dithering a bit about whether to use the improved TLF polling facilities (N out of up to 10, rather than 1 out of up to 5) to improve the polling part of the process, and if so, exactly how. The net result is that selecting 'UHYP16' got delayed long past the point where it could reasonably be called that...

And although I could in principle skip 'UHYP16' and just move on to selecting 'UHYP17', I'm again in the position of having too much other stuff I want or need to do more, partly as a result of a recent medical problem. So I'm afraid it's time for me to admit that I bit off more than I could chew with the idea of the UHYPxx yearly series, and abandon it. At least hopefully that will make it easier for me to catch up on the backlogs I've started accumulating on GDHYP and CHYP1...

Gengulphus

Thank you Gengulplus. UHP15 was indeed the post I was referring to. Hope you get better soon
Tuk020

[Hypster]

I think there may be a problem with that screening as Centrica fails to meet condition (5)?

Oh yes, well spotted CryptoPlankton. That metric was looking at % changes in the numeric yield rather than the dividends. I've swapped it for dividend-per-share rather than yield. And, because it's annualised that is also flawed because a run of high growth years followed by a cut last year could still show >5% annualised over the last five years. Therefore, I've added another filter, to pull out only those shares that have had five years of growth.