Another (probably stupid) question re "Great Owl" listings...

Old_Joe

OK, you folks may remember that I had previously mistakenly "inverted" the Great Owl (GO) "Return Group" ratings, resulting in total self-confusion, as usual. This question may reveal further shortcomings in my mental processing equipment, but please be kind and remember that those processors are well beyond the warranty period.

Here's the latest confusion... looking at FPNIX (FPA New Income) the GO charting shows FPNIX as a "5" for 20/10/5/and 3 yr periods, and as "4" for a 1 yr period... in the top two quintiles for those periods, as compared to the S&P generally.

Now M* shows the FPNIX "overall" score as 1*, as compared (presumably) to "other funds in it's class". By implication, it's more or less at the bottom of it's M* class, so there must be other funds in that class which M* feels are much better.

If that is so, then since FPNIX rates a "5" as compared by GO to the S&P, it would seem that entire class should also rate a GO "5" (compared to S&P), since M* claims that the entire class is equal to or better than FPNIX.

Now, something isn't adding up here. As usual, I suspect that I am doing something stupid, yes?

BTW, this situation is an outlier in my comparison of M* and GO ratings. In all of the other comparisons that I've made so far, the M* and GO ratingsw are pretty close, as you would expect. If GO thinks that a fund rates a "4" or "5" as compared to the S&P, then it's logical that M* will also place it near the top as compared to other funds "in it's class".

The spreadsheet version of the GO funds makes it very easy to add extra columns to the GO data, and I've added a M* rating column right next to the GO ratings, so this makes the comparison very easy (and in this case, dramatic) visually.

Edit/Add:

After posting the above, I revisited the definitions for a closer reading concerning the GO calculations. While it is true that the GO "Risk" calculations are in comparison to the S&P 500, this apparently does not hold true for the "Return" calculations:

Return Group

The score or ranking of a fund’s performance based on Martin Ratio relative to other funds in same investment category over same evaluation period (emphasis added). The evaluation periods are 1, 3, 5, 10, and 20 years, as applicable. Funds in the top 20 percentile are placed in return group 5, while those in bottom 20 percentile are in return group 1. MFO “Great Owl” designations are assigned to funds that have earned top performance rank for all evaluation periods 3 years or longer, as applicable.

Now, if I'm reading that correctly, both the GO comparisons and the M* comparisons are against "other funds in same investment category over same evaluation period". That would make sense, since it wouldn't be meaningful to compare, for example, a fixed Income fund's performance against the S&P as a benchmark.

Still, we're then left with a GO rating of 4 or 5 compared directly to a M* rating of 1. That doesn't seem to add up, either. Very confusing.

AndyJ

Well, M* classifies FPNIX as a 'non-traditional' bond fund, and it doesn't chart all that well against the average of that classification. But it also doesn't chart very well against other short term bond funds, which seem to be more its league. I guess I'd understand the M* rating, given the way they categorize it, and wonder how the GO rating came about; but like happens so often, classification can be destiny in the ratings game.

I looked at a long term chart, and never losing $ looks like it's probably an objective. That must mean something for classification ?!?

mrdarcey

I had problems with this too. Someone can correct me if I'm wrong, but this is basically one of those times where M*'s peer groupings create confusion. M* recently moved FPNIX to the "Nontraditional Bond/Flexible Income" grouping, after FPNIX stated it's appreciation goal to be CPI + 1%. Its now compared with funds like LSBRX and PUBDX, which had been on fire over the past five years.

But FPNIX is super conservative and has been less concerned with total return than with capital preservation. From FPA's policy statement fpafunds.com/docs/fpa-new-income-fund-information/2011-03-the-fpa-absolute-fixed-income-policy-statement.pdf?sfvrsn=2:

The FPA Absolute Fixed Income Strategy (including FPA New Income, Inc.) is one of the longest standing fixed income strategies in the USA. We have a defined investment philosophy and process and have executed against it for almost twenty eight years. At various times during this period, both aggressiveness and caution have been demonstrated in the selection of longer term and credit sensitive investments. Only when we felt that we were being more than adequately compensated for the potential risk of loss did we become aggressive.

For the past eight years, our investment strategy has been one of caution since we believed that risk of loss was too great and thus, we focused on capital preservation first, income generation second and capital appreciation last.

M* says as much in their Fund Analysis as well. FPNIX is down to around 1.4 yrs avg. duration, and by rule can't hold more than 25% junk bonds. Basically they're a fish out of water when being compared to other Multisector/Non-trad funds. A closer comparison might be either Short Term funds like JASBX or SCLDX, or some grouping of conservative funds that fall outside of traditional definitions like PYGSX or RPHYX.

Mona

Reply to @mrdarcey:

M* says as much in their Fund Analysis as well. FPNIX is down to around 1.4 yrs avg. duration, and by rule can't hold more than 25% junk bonds. Basically they're a fish out of water when being compared to other Multisector/Non-trad funds

Thanks for the information. I get the duration and that FPNIX can't hold more thab 25% in junk bonds. Where I continue to be lost is what type of bonds does the fund hold?

Mona

catch22

Hi Mona,

I have not looked at the fund prospectus; but this link shows Fido's breakdown, as of 6-30-2013.

Old_Joe

Well, it looks like the discrepancies may be due to what is termed "category drift". To make a longish story short, apparently FPNIX is a rather strange duck... different things to different raters. Not bad, just different. That actually makes it pretty interesting, at least to me.

Thanks all!

Charles

Reply to @Old_Joe: Please note also that I believe M* system uses Sharpe Ratio when ranking the risk adjusted returns and calculating number of stars, whereas MFO uses Martin Ratio.

In other words, M* uses return relative to volatility index (Standard Deviation, STDEV), but MFO uses return relative to drawdown index (UI, Ulcer Index).

And, fact is, FPNIX has pretty high Sharpe Ratios but even higher (relatively speaking) Martin Ratios. So, without digging further (for now) I'd argue it's classic case of "good" versus "bad" volatility.

Bottomline: FPNIX has delivered increasing returns with little drawdown, thereby elevating its ranking on MFO system.

Hope that helps. Appreciate feedback, as always, and opportunity to improve.

Old_Joe

Reply to @Charles: If any improving needs to be done, I suspect it's on my end. Thanks much... have a great trip. You are, at the moment, not very far north of our weekend place at Guerneville, on the Russian River. If by any chance you intend to come down our way on a Saturday, Sunday, or Monday send me a MFO message and perhaps we can get together for dinner or some such.

Investor

Reply to @Charles: Here is M* methodology:

http://corporate.morningstar.com/bl/documents/MethodologyDocuments/MethodologyPapers/MorningstarFundRating_Methodology.pdf

It even gives formulations...

Charles

Reply to @Investor: Thanks man. M* methodology uses Sharpe, like I remembered. I think documenting their methodology of assigning stars for both funds and stocks is something M* does very well.

Charles

Reply to @Old_Joe: You sir are a jewel.

Mona

Reply to @catch22:

Hi catch,

Thanks.

I did not realize Fidelity has such a nice layout.

Mona

catch22

Hi Mona,
I should have (been too busy here) noted the page link as a decent place to review funds. I like the layout, too.This is a fairly new function/layout at Fido and the nice part, too; is that anyone may use the area. The other "tabs" are useful, too.
Keep this Fido page on your favorites list, eh?
Take care,
Catch

Old_Joe

Reply to @catch22: Thanks Catch- I had just added it to my permanent browser tabs.