P2P Lending Webinar Now Available for Viewing

Last week I conducted a peer to peer lending investor Q&A webinar. As promised I recorded it and you can watch it below. It is one hour and five minutes long. This is the raw footage – I have not edited it in any way. I hope you find it interesting and educational.

For the best video quality be sure to click on the HD button above to watch this video in high definition – it makes the screen shots much easier to read.

Notify of
Newest Most Voted
Inline Feedbacks
View all comments
Dan B
Dan B
Jul. 19, 2011 4:27 pm

I know that you Peter use Lendstats extensively during this webinar as well as in general. However I’ve been a bit underwhelmed with the enormous disparities that have come up when looking at my Prosper account in the 4-5 months that my account has been in existence. Currently, my ROI through Lenstats is at 3.2%. An email I received from Prosper today estimates my ROI at 17.5%. This is an extreme example to be sure but variances of over %% have been a consistent feature in my experience. Since I haven’t made an investment since mid May, these discrepancies cannot be attributed to “new note” purchases. I’m sure that there will be no shortage of explanations for this disparity……………but in my opinion, the fact that this has to to be explained in some fashion does in a severe way undermine the use/necessity/function of the site in the first place. No offence Ken, but that’s just how I see it.

Jul. 19, 2011 5:08 pm

The difference is that I penalize for loans that are between 1 and 120 days late. Prosper (and LendingClub) do not. If you disagree with the way I penalize for late loans then you can set the “loss factors” to zero for the late loans and surely then the lendstats ROI will be very close to Prosper’s.

Dan B
Dan B
Jul. 19, 2011 6:31 pm

Saying that something is an “estimate” is all very nice & fine. But there comes a point that an “estimate” loses most of its meaning when it differs too much from “reality”. To have a difference of 13-14% between the 2 sources is quite astonishing. I don’t see any reason why an investment that is 5 yrs old now cannot come up with an easy & universally accepted method to report something as crucial as ROI. To have “estimates” that are all over the place is the best we can do?

Jul. 19, 2011 7:01 pm

Dan, if you set the late loss factors at LendStats to zero and the default loss factor to 1, then what is the difference between your Prosper ROI and your LendStats ROI ? If you want to criticize lendstats, that is the fair comparison to make.

Dan B
Dan B
Jul. 19, 2011 8:25 pm

You’re missing my point. I’m not really criticizing Lendstats. I’m critisizing the simple fact that this industry can’t even agree on how to report something as basic as returns.

Here’s the question & multiple answers as it stands right now. Hey Dan, what return are you getting on your Prosper investment?

The answer………….Well Prosper says I’m getting 17.5%. But Lenstats, the largest & most respected independent site says I’m getting 3.2%. Of course if you go there & manually adjust the late loss factors to zero & etc etc. Of course you can also run an Excel & come with another number. I mean come on, why all this complication? Just a straight simple answer.

Brian B
Brian B
Jul. 19, 2011 9:00 pm

There is no factually correct way to calculate your ROI (or even your account balance for that matter), so your “straight simple universally accepted answer” does not exist.

This results because there are no straight simple universally accepted answers to the following questions:
-should notes that are late have their principal discounted, and if so, by how much? (Lendstats says ‘yes’ to discounting)
-should idle cash in your account be incorporated into ROI calculations? (prosper says ‘no’ to incorporating idle cash)

But once you answer these questions for yourself – then you will know which of the calculations you should use.

It might actually be more accurate that instead of saying there is no straight simple answer, that there are many straight simple answers, and you can use whichever one you want.

Dan B
Dan B
Jul. 19, 2011 9:33 pm

Thank you Brian, though I do know all that. Of course the complications & interpretations are much deeper than that aren’t they. For example, I understand that Lendstats chooses not to return a note to current even after it actually returns to being current. Of course one can manually compensate for that…………….but we’ve already covered all that, haven’t we.

But putting all that aside, I’m not asking for a definitive right or wrong way answer here. I’m asking for an agreed upon method, that’s all.

James L
James L
Jul. 20, 2011 8:06 am

An agreed-upon method may in the works! The recent GAO report mentioned the possibility of creating a reporting standard for the statistics in social lending.

“Staff from SEC’s Division of
Corporation Finance said that SEC could potentially encourage or
require Prosper and LendingClub to report in the same way, either
through the SEC staff comment process for the companies’ financial
disclosures or by proposing a rule that would require them to use
consistent reporting methods.”

I’m interested to see if this could solve the discrepancies between the many formulas to generate ROI/rate of default statistics.

Jul. 20, 2011 10:41 am

LendStats does not have access to the trading platform information, and can therefore never be accurate. In my account it shows ROI 23% because of 3 late notes, but in reality I sold those notes long ago.

Jul. 20, 2011 11:41 am

LendStats does return Late notes to current status. There are 3 cases I need to explain where it appears that LendStats does/has not.

1)Prosper has started changing late notes to current while the payments are still processing. The payment, however, does not show up in the data export until it has gone through (as it should). Therefore, in this case it often appears as though LendStats does not return loans to current, but the reality is that they are not yet current.

So in this case, IMO LendStats is correct and Prosper is wrong.

2) At least 2 times, for very long extended periods of time Prosper has delayed some information in its data export. Why they did this, I do not know (but I have my theories). Each time I informed them about it, they promptly fixed it. These delays caused what seemed like errors at LendStats. The most recent of which I informed them about 1 week ago.

In this case the fault is entirely on Prosper, not LendStats.

3) Prosper does not have <15 day late info in their export, but they do have other info which I can use to figure out if a loan is <15 days late. However, when I do this, if there is a payment during that 15 day span, lendStats may not detect it due to lack of information in the data export. I have realized though that maybe I can detect that payment, but I haven't done the programming yet. It could be something difficult to program and debug and I'm afraid to get started with that due to the amount of (unpaid) time it could require. However, if the next payment is made, then the loan is returned to current.

In this case, I have to say I am guilty. Sometimes the <15 day late loans (and only the <15 day late loans) will remain in that state for the entire month. However, I do not think that this is such a bad error because the probability of a cured <15 day late loan going late again is elevated.

As far as I'm concerned, the proof is in the pudding. I like LendStats, I'm using it and I'm sharing it with others (for free). Since I created it I've been getting returns at Prosper of around 20% on about 1000 loans. Without LendStats, my returns would be much much lower and that is a fact.

Jul. 20, 2011 12:04 pm

Yes, that is also a problem. The export file only shows loans, payments and bids on the Prosper platform. No folio data is available.

I too actually have not had a default yet with the post-SEC registration Prosper platform because I sold off my 6 loans which have since defaulted a long long time ago.

Jul. 20, 2011 12:23 pm

Peter / Ken,

Wanted to get your thoughts on what effect worth-blanket2 will have on near-term / long-term ROIs, with his investments of $5000-$7500 per note?

Here are my thoughts: Before worth-blanket2 began, there seems to be some “equilibrium” of supply/demand, where statistically “good” notes would fund in under an hour, while “bad” notes would fail to reach 70% funded (in other words, I would assume that notes that fund in an hour or a day have fairly low default rates). This accounted for the recent 10.6% ROI (per Prosper). With more notes funding, that otherwise may not have funded, I think that this advertised ROI will fall by .3 or .4% over the next 6 months.

Your thoughts?

Jul. 20, 2011 12:38 pm


Since Prosper’s 10.6% is based on loans that are 10 months old, WB2 will be having no effect on those calcs for a long time. However, WB2 has funded what look like very risky loans and he could very well drive down overall ROI. Luckily, it looks like he’s now taking a slighly more conservative approach and staying away from the HR’s. My guess is that he has improved his standards and hopefully his loans made from now on won’t have such a negative impact.

Peter, you should check those formerly Late loans and see if the payments are “processing” or if the payments already went through. If they are still “processing” there is a good chance the payments will fail and the loans will go Late again.

James L
James L
Jul. 20, 2011 1:24 pm


Prosper has just recently increased interest rates for A and B borrowers.
I’m going to guess your webinar had some influence 🙂

Source: https://www.doughroller.net/investing/prosper-raises-rates/

Brian B
Brian B
Jul. 20, 2011 7:00 pm

, yes that is 1 way to calculate ROI. But its not the only correct way, there are many others.

My ROI display preferences:
Lendstats Method > Lending Club Method > actual return

Jul. 21, 2011 4:45 am

Regarding the @XIRR spreadsheet function, you can reverse the polarity of the numbers, and still come out with the same exact result. For most people, it is probably more intuitive to enter deposits as a positive number and withdrawals as a negative.

On the last question, just because the bank or loan processor goes bankrupt doesn’t mean the debt obligation goes away. I’m sure some were excited when Wachovia and Washington Mutual went under, but the loans were assumed by the banks that took them over, so nothing changed. I suppose the P2P backup loan processor could decide to start charging 100% commission or a bankruptcy judge could freeze loan payments as they come in, but I would hope regulators and a little common sense would prevent that from happening.