What's This "Best Note" Selection Business Anyway?

[Rob L]

The loan issuer may want to optimize its model for consistency, not absolute returns.

Outstanding. LC certainly has every incentive to do just that and, if true, would explain my confusedness.
Maybe your posit isn't true.  Could be some other divergence between my goals and theirs but I'm a much happier camper now.
Thanks!

Note: Spell checker is cool with the word confusedness. Guess there's enough of us around to put it into the dictionary.

[Dennis]

Rob, you post a great question with this topic, one that has me really thinking.  I haven't been on these boards for about 2 months now because the posts were becoming way too social and not very informative.  But now you pose this great question....

I admit I'm one of those "best notes" people, in that I have my own criteria or "subset" as you put it, in choosing notes.  The challenge then, for me, is to beat the averages which I have consistently done for 2 1/2 years now.  This is a hobby for me, so I've mostly enjoyed the note selection process and I still do most of it manually.  I have been using filters at LC somewhat the last several months though as it has become increasing difficult to get those "best" notes if you're not fast enough.  But I use no filters at Prosper, and still hand pick (after using a general filter) all my notes at LC.

So personally I do think there is a subset within note grades (that's just my opinion) that can outperform the average in that grade.  I have nowhere near the skill set of many here who can run numbers, accurately configure probabilities, or create API's or other software, but I still do okay.

I have 3 P2P accounts, and even with the many defaults I get because of my high risk exposure, my current combined weighted average return for those accounts is currently 14.72%.  I've hovered around 15% for the last year and am currently at about the lowest return rate I've been at in 2 1/2 years.  The first notes I purchased will start paying off in the next 6 months (for those that go full term), and my 3 year experiment with P2P should yield some interesting results.  I am impressed with it so far.

Maybe I've been lucky, maybe I've stumbled on some skillset I don't know I have yet, or who knows what, but I will continue doing what I'm doing as long as I continue to get these results, which means that I believe there is a subset of notes that will outperform the averages. 

All that said, I don't like all the defaults I get, I hate when borrowers make less than 10 payments, and in some cases they make only 1 or 2 payments and then pay off the note - those things waste my time and frustrate me, but I've learned that it's just part of the game and you have to accept it.

Again, great question........

     

[Rob L]

Rob, you post a great question with this topic, one that has me really thinking.  I haven't been on these boards for about 2 months now because the posts were becoming way to social and not very informative.  But now you pose this great question....   

Welcome back Dennis. My problem was a conundrum. From all evidence it appeared as if best loan selection was possible and a universally believed truism. I just didn't see how that could be, though I conceded it probably was.

Emmanuel's post was a real eye opener. The goals of LC's model and my personal investment goals are just not the same. It's obvious in hindsight but that hadn't occurred to me. Duh.. Where they diverge in my flavor I (we) have the opportunity tho select better notes.

I don't pretend to understand this very well, but I'm much more comfortable now with the possibility (reality) that better note selection is possible. The good news is that you're probably not just lucky but you're doing something right.

[Randawl]

From all evidence it appeared as if best loan selection was possible and a universally believed truism. I just didn't see how that could be, though I conceded it probably was.

I don't believe it's a universal truism that the "best" notes are always known ahead of time nor that the ones fully invested quickest are the "best."  Filtering is a double-edged sword and algo's/machine learning is imperfect as well.  They are useful tools but often over relied upon.

The goals of LC's model and my personal investment goals are just not the same.

That's often what we fall back on when we try to figure out why LC made another incongruent move.     

[Dennis]

Emmanuel's post was a real eye opener. The goals of LC's model and my personal investment goals are just not the same. It's obvious in hindsight but that hadn't occurred to me. Duh.. Where they diverge in my flavor I (we) have the opportunity tho select better notes.

The really great thing about this board is that it allows the convergence of thought from many different points of view.  Sometimes what's right in front of you will lay hidden until someone coming from a different angle sheds light on it.  I've been oblivious to the obvious so many times in my life that I stopped counting.  It doesn't mean I'm stupid, only that I was looking at something from the wrong angle.  This board has helped me so many times........     

[Fred]

We do have the opportunity to select the best E3's (or whatever) loans.

Looks like you found an answer to your question.

[rawraw]

Rob, you post a great question with this topic, one that has me really thinking.  I haven't been on these boards for about 2 months now because the posts were becoming way too social and not very informative.  But now you pose this great question....

I admit I'm one of those "best notes" people, in that I have my own criteria or "subset" as you put it, in choosing notes.  The challenge then, for me, is to beat the averages which I have consistently done for 2 1/2 years now.  This is a hobby for me, so I've mostly enjoyed the note selection process and I still do most of it manually.  I have been using filters at LC somewhat the last several months though as it has become increasing difficult to get those "best" notes if you're not fast enough.  But I use no filters at Prosper, and still hand pick (after using a general filter) all my notes at LC.

So personally I do think there is a subset within note grades (that's just my opinion) that can outperform the average in that grade.  I have nowhere near the skill set of many here who can run numbers, accurately configure probabilities, or create API's or other software, but I still do okay.

I have 3 P2P accounts, and even with the many defaults I get because of my high risk exposure, my current combined weighted average return for those accounts is currently 14.72%.  I've hovered around 15% for the last year and am currently at about the lowest return rate I've been at in 2 1/2 years.  The first notes I purchased will start paying off in the next 6 months (for those that go full term), and my 3 year experiment with P2P should yield some interesting results.  I am impressed with it so far.

Maybe I've been lucky, maybe I've stumbled on some skillset I don't know I have yet, or who knows what, but I will continue doing what I'm doing as long as I continue to get these results, which means that I believe there is a subset of notes that will outperform the averages. 

All that said, I don't like all the defaults I get, I hate when borrowers make less than 10 payments, and in some cases they make only 1 or 2 payments and then pay off the note - those things waste my time and frustrate me, but I've learned that it's just part of the game and you have to accept it.

Again, great question........

   

Are you continually adding new money?  If so, this could keep your returns high for a bit until the average age increases.

[Rob L]

Are you continually adding new money?  If so, this could keep your returns high for a bit until the average age increases.

No, I'm not even re-investing payments received; just letting all my notes run off. My question wasn't prompted by poor performance and I understand your comment regarding the returns of young notes. My account's about 6 months old and doing fine (though not multiple sigmas above the curve). My question was spurred simply by a fundamental lack of understanding of "why", and the possible similarity of arguments for and against technical analysis of stocks, random walk, etc.

My account is traditional IRA. When LC changed the rules after I joined to prohibit my use of Folio I was robbed of the option to liquidate my account if I chose.
LC has said they plan to address this in the future, but it isn't a priority. Until they do I'll continue to let the account run off and invest the dollars returned  elsewhere.

I am planning to open a non-IRA account and basically start over. For me it's an interesting, enjoyable and profitable hobby so I plan to stay in the game.

[dontvote]

Evaluating credit like this seems more like fundamental stock analysis than technical stock analysis. I would consider the analogy of picking notes out of LC's credit model being more like analyst rankings at your brokerage. With a bit of judgement and significant diligent effort, you can probably do better than the analyst (or at least pick a subset of his picks that fit with your investment scheme).  You can apply two different models to the same data and get very different results. The model doesn't have to be filter results based or algorithmic, it's just a framework you can test.

[Emmanuel]

Emmanuel's post was a real eye opener. The goals of LC's model and my personal investment goals are just not the same. It's obvious in hindsight but that hadn't occurred to me. Duh.. Where they diverge in my flavor I (we) have the opportunity tho select better notes.

You're too kind! That's one explanation, but keep in mind I may be totally wrong ;-)
In any case, we plan to do some serious statistical analysis in the near future, so hopefully it will bring a bit more light

[Dennis]

Are you continually adding new money?  If so, this could keep your returns high for a bit until the average age increases.

rawraw, I think you were directing that question towards me: 

It's been about a year since I put any new money into any of my 3 accounts.  I personally have found it near impossible to add new money now that there's so much competition for notes.  I've been struggling to just keep up with reinvestment, but have fallen behind on that now, through the holidays.  I have a good deal of cash sitting in all my accounts presently.  So my portfolios therefore are pretty seasoned.  Like I said, I'm 2 1/2 years into this now, and my return is at about the lowest it's been in that time.  I currently hover around 15%, but it was much higher a year ago when many of my notes were newer as you suggest. 

[Bohb Daishi]

I would consider the analogy of picking notes out of LC's credit model being more like analyst rankings at your brokerage. With a bit of judgement and significant diligent effort, you can probably do better than the analyst (or at least pick a subset of his picks that fit with your investment scheme).

It's not hard to beat analysts. In fact, most don't even know what the heck they are talking about. If they were smart enough to accurately predict where a stock would go, they would be fund managers. I've read many research notes and I can tell you, most of these guys have absolutely no idea what the company actually does. They have an even worse idea of how to value the company.

Remember, these are the same people who said the housing market would always go up, gold would always go up, subprime mortgages are "AAA", dotcom tech stocks would always go up, and that zero-revenue social media companies are worth billions of dollars.


What I'm trying to say is, there are always going to be very big arbitrage opportunities in "one size fits all" pricing models, like what LendingClub and stock analysts use.

[Randawl]

What I'm trying to say is, there are always going to be very big arbitrage opportunities in "one size fits all" pricing models, like what LendingClub and stock analysts use.

That's a great way to put it

[Joleran]

It's not hard to beat analysts. In fact, most don't even know what the heck they are talking about. If they were smart enough to accurately predict where a stock would go, they would be fund managers.

If they can repeatedly and accurately pick where any stock or set of stocks would go, they would be billionaires.  The number of funds that have been able to produce any alpha whatsoever on a consistent basis for more than 10 years can be counted on one hand.

Think about it like this - how many funds underperform their index every year?  If you have a thousand funds and you assume it's completely random, you might expect 500 outperform the first year, of which 250 the second year, 125 the third year, 62 the fourth year, 31 the fifth year, 16 the sixth year, 8 the seventh year, 4 the eighth year, 2 the ninth year, and 1 the tenth year.  In reality though, 80% of fund managers on average underperform their index so the odds are worse.

[Rob L]

You're too kind! That's one explanation, but keep in mind I may be totally wrong ;-)

Of course your example may be wrong, but there are any number of other similar divergences of goals that are possible. I just didn't see that.
For the sake or argument lets say your argument is right. Does that mean some borrowers get better deals than others simply for the sake of consistency optimization?
The borrowers that get the bad deals relatively speaking are the loans we seek. Or, did I miss the point completely?