Two New P2P Lending Statistics Sites Launch

Regular readers of this blog will know that I am a huge fan of the statistics site Lendstats. I have based my entire p2p investment strategy on queries that I have run on that site. Now, there are not one, but two new p2p lending statistics sites that have launched recently.

Nickel Steamroller p2p lending statistics site

The person behind Nickel Steamroller is a big fan of Lendstats, he says that site inspired him to create his own statistics site. In some ways Nickel Steamroller provides some of the same data as Lendstats does but he has added a few new features as well. Right now it only works with Lending Club data but there are plans in the works to cover Prosper as well. In the interests of full disclosure I should say that the owner of Nickel Steamroller contacted me a few weeks ago to give some feedback so I have provided several suggestions to him, many of which have been implemented on his site.

First let’s take a look at the core statistics. The Get Started tab provides an overview of the ROI for different criteria, but what is more interesting is the Return Rates function. This is where you can run your own queries against the Lending Club data. It provides a pretty simple design that is less cluttered than the Lendstats site. It doesn’t have every data point but it hits most of the important ones. It provides an estimated ROI, the default rate, a return-to-risk ratio (which is just ROI/default rate) and the total number of loans. Nickel Steamroller is updated every day with new Lending Club data.

New Tools for P2P Investors

There are two new tools on Nickel Steamroller that I think many Lending Club investors will find useful. The first is the Analyze Your Portfolio option. If you download your entire Lending Club portfolio as a CSV file and then upload it here Nickel Steamroller will analyze it and make suggestions. Some of the suggestions you may or may not want to take but I found it useful in identifying places where I had accidentally invested twice in the same note. It highlights these notes for you.

The other new tool that is pretty cool is the saved searches and email alerts. If you create an account on Nickel Steamroller you can save your searches and then sign up for email alerts. Then every morning Nickel Steamroller will send you an email with new loans that have recently come on the platform that meet your search criteria. You can click on the link to these loans in the email and invest in just a few clicks.

Smart peer lending p2p statistics

The other new statistics site is Smart Peer Lending which is also focused just on Lending Club statistics for now. It is not as extensive as Nickel Steamroller but it does provide a loan analyzer tool. This tool, while useful, is a little bit cumbersome to use but it does hit pretty much every data point that Lending Club provides. There is also a tab for Lending Club statistics that summarizes the entire Lending Club loan database into different criteria, although some of the data here looks like it has some problems.

One useful thing that Smart Peer Lending has done is include an area where you can just list a bunch of loan numbers and it will calculate the ROI on those loans. I downloaded all the loans in my new IRA, copied and pasted the loan numbers and it calculated my ROI. It was about three percentage points less than my Lending Club NAR, which is likely a better estimated ROI than the Lending Club NAR on such a young portfolio (my average loan age in this account is just 2 months).

Comparing ROI Calculations

The most important piece of information for p2p investors is the ROI calculation. I took a relatively simple filtering criteria so I could compare the estimated ROI on both these new sites and Lendstats. I chose all D and E grade loans with a maximum number of inquiries of zero. Here is the ROI comparison:

Lendstats – 8.44%
Nickel Steamroller – 9.96%
Smart Peer Lending – 7.28%

Obviously there are some big differences here. Part of that can be explained by the loss rates. Nickel Steamroller uses loss rates that are recommended by Lending Club whereas Lendstats uses its own (less optimistic) loss rates, which causes a lower estimated ROI. Some of the reason that Smart Peer Lending is lower is because is sets every loan to $1,000 and it may use different loss rates as well, I am not sure. My last email to the owner of that site has not been answered as of this writing.

My Take on These New Statistics Sites

While Ken from Lendstats might not be thrilled with the competition I think for p2p investors this can only be a good thing. These two new sites are not quite ready to supplant Lendstats as the p2p lending statistics standard, but they do provide an alternative for investors. Both developers have indicated they want to continue improving their site. Let them know what you think and what they need to work on in the comments.

Notify of
Newest Most Voted
Inline Feedbacks
View all comments
Matthew Paulson (P2P Lending News)
Aug. 19, 2011 1:29 pm

Look forward to seeing these new tools develop 🙂

Aug. 19, 2011 1:34 pm

I’ve tried 4 times to register. It keeps kicking me back to putting my password???

Aug. 19, 2011 1:47 pm

Jim, try now. I migrated to a new host yesterday and a database schema change didn’t make it over (oops!). A feature that I added yesterday was the ability to turn your alerts on and off (on by default) so if you want to keep your saved searches and not be notified, you can now do this.

Thanks for checking it out. I hope to hear your feedback.

Aug. 19, 2011 5:47 pm

Just came across a new hickup in Lending Club’s site. If you invested in a loan and the same member takes out another loan in the future, Lending Club will flag it with the “already invested in” symbol. I noticed this because I saw a few times that symbol on a loan I knew I had not invested in. When contacting LC they explained that their system will identify ‘already invested in” notes by member rather than by note.

Now, not enough Lending Club does not provide Lending Club history for repeat borrowers, but if you ever lent this member before they will also flag the second loan as alread invested, preventing you from investing in it! now thats really screwed up…

Dan B
Dan B
Aug. 19, 2011 6:58 pm

Well it doesn’t really “prevent” you from investing in it does it? I thought that it justs lets you know that you have already invested in it previously.

Aug. 19, 2011 8:30 pm

It makes you think you invested already in the loan, maybe you have a photographical memory and remember all loans you invested in… I only noticed this because the loan was in a grade I don’t invest in. Also it will filter it out in the search if you choose to hide loans you invested in already which most investors probably do.

Dan B
Dan B
Aug. 19, 2011 10:56 pm

Moe…………I understand what you’re saying & thank you for pointing that out. I’m guessing that it’ll probably filter it out ONLY if the other loan you invested in before is still active. Is that correct?

Aug. 20, 2011 9:07 am

It’s very interesting. Can it be set up to notify you of a specific type of loan when one becomes available? I did take the analyze advice and listed some for sell that was recommended. We’ll see where that goes. Thanks

Aug. 20, 2011 4:41 pm

Dan, I’m not sure about that and LC didn’t tell me. It’s not realy this particularly that bothers me, it’s just their whole system is unprofessional and gives the impression of something run manualy. If you notice that on the day a payment is due, the acrrued interest on those notes will go down to 0 and on the day the payment goes through this happens again. Also I once had a transfer that didn’t go through, but LC added the money to my account. I reported it to them but they said that they see it went through… It took them half a year to see that it didn’t go through and I got an email from them. There are also a few other minor problems in their website such as the sorting function. I’m no expert in banking systems but I never saw such things on any decent website.

Aug. 22, 2011 10:39 am

I wish good luck to both of these sites, but beware. A website can become a blackhole that makes time disappear without a trace, lol.

I do have one question though. I have done a little searching on both sites looking for how ROI is calculated, but I didn’t find anything. How are the ROIs calculated?

Aug. 22, 2011 10:49 am

I am using your calculation, which is in my code roughly as:

$return = @number_format(($total_gain / ($interest_over_rate + $total_gain)) * 100, 2);

As Peter mentioned, I used more optimistic loss factors than you’re code assumes (I took the numbers from LC’s publication on default loss and recovery). I also do not expose this as an option to the user to change like

I need to make this equation into an image and then I am going to integrate it as well as show the default values I am using on the site.

You are right on the time commitment. I’m hoping that my automation take s care of it after my initial time commitment which is probably close to 100 hours now.

Aug. 22, 2011 11:48 am


Since you are using my equation, I do think that the polite thing to do would be to reference it as such. Although the equation seems simple, it’s development was not. It took me many months to develop that equation. I had a completely different equation in the beginning and I continually tweaked it to improve it’s performance, until one day….. wala!! The tweaks I had been making simplified and cancelled out and the equation which I use and you use and which I assume is now the excepted standard (even though certain significant parties did not want to except it) for ROI calculations appeared.

Aug. 22, 2011 12:51 pm


I’ve thanked you on my blog for providing the inspiration for my work last week. I can certainly extend the credits on my site to mention your equation as well.

Aug. 22, 2011 12:59 pm

Thank you Michael, I appreciate it.

Aug. 22, 2011 4:43 pm

: Since you are saying that LendStats is “more negative”. Are you implying that Lending Club’s loss factors are more accurate, than LendStats’s?

Charlie H
Charlie H
Aug. 23, 2011 7:32 am

@ RJ

KenL uses what he believes to be more realistic Loss Factors then what Lending Club publishes. The differences how you calculate the loss.

Some one with experiance on the Folio platform could tell you how the Market prices “distressed” loans and use the average discount as the Loss Factor. Assuming the Folio Market Place prices loans rationally (which is a big assumption right now) this would be more accurate still.
(Mark to Market pricing vs Mark to Model pricing)

Aug. 23, 2011 9:56 am

@Charlie and RJ and Michael,

Michael is free to use whatever loss factors he wishes, it makes no difference to me. It is his website, not mine (even though he uses my ROI equation). I do not think I need to justify the loss factors that I use, but I’ll do it for the sake of debate. My loss factors are based on my own experience as a p2p lender (4+ years, ~4000 loans) and as a number cruncher (~2 years, 2000+ hours maybe more). And guess what folks, there ain’t no substitute for experience :). If you need statistical proof, all you need to do is look at LendStats and compare returns from 2010 to 2011. If indeed my loss factors are too harsh, as time goes on my calculated ROI’s would go up from year to year (or quarter to quarter) but actually the ROI’s go down as you go back, and they go down significantly. This implies that maybe my loss factors are actually not harsh enough.

And it also needs to be pointed out that loss factors that I use ARE NOT the fraction of ALL late loans that default. They ARE a fraction of CURRENTLY late loans that default. There is a difference in those fractions of (ALL or CURRENTLY) late loans that default and if that difference is not understood then any further debate on the subject will be useless.

All that said, my loss factors are not set in stone and it is possible that over time it will become evident that they are too harsh, but the contrary might happen as well. But as of now, I see no significant evidence indicating that I should make any significant changes in either direction to my loss factors.

Aug. 23, 2011 10:10 am

After rereading my post I realize I made some confusing statements. Let me try to clear them up.

If my loss factors are too harsh, ROI’s would go up as the loans age because the loss factors have a larger effect on the ROI’s of a younger loan set and less of an effect on loans as they get closer to maturity. However if we compare the younger 2011 loans to the older 2010 loans, we can see that the older loans have a much lower ROI. This implies that my loss factors may not be harsh enough.

Aug. 23, 2011 11:33 am

@ All: Dont mean to offend anyone, I am new and started a couple of months ago. So please dont take anything personal.

@ Peter: When I read your article I did NOT see what the loss factor were that you used? What is the LC recommended loss factor? and What is the Lendstats loss recommended…(is it his default?)

My only point is that I was trying to understand your comparision. Perhaps maybe you had a poor choice of words (less optitmistic). I think all lenders want “reliability”?……

As a side note, my account @ LC is about 2 months old and evertime I log into I see the words “Net Annualized Return” and I used to be pretty happy until…I realized that I have only been a member for 2 months. This number doesnt tell me anything and I wish they wouldn’t show it…. it will take atleast nine months, and it does not factor fofilo. On the flip side when I choose a basket of loans to invest in, it gives me a weighted average interest rate and then some “loss factor”. So I imagine first number numbers is created by the marketing department and the latter is probably for discloure…but I am not sure.

So Peter when you state that you are using a Recommended Loss factor…what is the number? and if you know what is it based on?

Charlie H
Charlie H
Aug. 24, 2011 6:58 am

“My loss factors are based on my own experience as a p2p lender (4+ years, ~4000 loans) and as a number cruncher (~2 years, 2000+ hours maybe more). And guess what folks, there ain’t no substitute for experience . If you need statistical proof, all you need to do is look at LendStats and compare returns from 2010 to 2011.”

Which agree’s with my statement that:
“Ken uses what he feels are more realistic Loss Factors”

Currently Late is a good predictor of default
Ever Late (a loan that is now current but was late at some time in the past) should also be a good predictor of default, but I don’t know that to be true.
On Payment plan should also be a good predictor of future default.

By predictor of default I mean the likelyhood of a Late, Ever Late, or On Payment Plan eventually default is higher then an otherwise identical loan that has Never Been Late.

Eventually if one believes in the Rational Market theory, Folio should do a bet job of pricing distressed loans dynamically then a model does because the model is static and does not take into account changing economic conditions. Example the default rate now should be higher then a default rate in an economy growing at 4% a year with a unemployment rate <6%. That would cause loans that are distressed to be priced differently because the likelyhood of default would be lowered.

Right now Folio is not a rational market place. I see people doing silly things all the time. In agragate, eventual it should become more rational.

Aug. 24, 2011 9:42 am

Thanks Charlie, I do want to do an “ever late” analysis eventually and also a on payment plan analysis, but it all takes time. Hopefully soon I can get to work on it.

As long as there are new investors jumping into folio, it will continue to be irrational. Maybe in a few years, if newbies will be a small fraction then maybe rationality will start to take hold.

Aug. 24, 2011 11:59 am

: I was just getting curious, bc they are using the same formulas. Ken as stated that his is based on experience. Has Michael stated what his is based on?

@Michael: What is your rational for picking less aggressive loss factors other than computing a higher ROI than LendStats?

Aug. 24, 2011 12:16 pm


Ultimately what you are looking at with Ken’s equation is I=Prt (I=interest, Principle, R = rate, t = time) expressed in terms of ROI. What Ken does is use r/I for P which is ingenious (t is assumed to be 1). Up to that point we use the same formula. Where we differ is with loss factors.

I take mine from LCs website:

It’s important note that these are most likely optimistic (blame the marketing department) since they only capture a small amount of operating history (6 months). My numbers are not based on experience as I have only been investing for about 2.5 months now. But that is where my data comes from.

Aug. 24, 2011 12:22 pm

I just looked in the library, these the the rates I am using:

Default: 1.00
Charged Off: 0.97
Late 16-30: 0.45
Late 31-120: 0.25
In grace: 0.20

I am working on a new rev of the site that will include this information in a more transparent manner.

Charlie H
Charlie H
Aug. 24, 2011 2:02 pm

The Crazy thing about this KenL is that people who work in the unsecured credit industry already have a real good idea of what the loss factor has traditionally been. “We” are mostly reinventing the wheel.

Once an account goes past due, what are the chances that it will become current?
Once an account is 15-30 days late what is the chance that it become current?
Once an account if 30-120 days late what is the chance that it will become current?
I am sure this has already been modeled by the large Unsecured Consumer Credit companies. Its most likely properitary and not something one could find in a 10K filing however. 🙂

Aug. 24, 2011 8:28 pm


I agree, we are re-inventing the wheel, but who knows, maybe the wheel is a little different for us.

But keep in mind, the loss factors do not represent the fraction of all late loans that default. Let me give an example, a loan that cures on the 16th day of being late will only be visible as 15-30 late for one day. But a loan that goes through the entire 15-30 day cycle as late will be visible as late for 15 days, therefore it will be 14 times more likely to be seen as late as the loan that cures on day 16. For this reason Loss Factors as lendstats uses them are a different animal than those factors that LC has on their stats page. I’ve explained this to LC, and they understood initially, but it appears that they have forgotten. Oh well, such is life.

Aug. 24, 2011 8:32 pm


At least someone here is able to recognize the genius of my equation,


Charlie H
Charlie H
Aug. 25, 2011 8:38 am

Thanks for all the work you do KenL!

Aug. 25, 2011 12:19 pm

Danke, I guess experience is real. Seems like there are a few independent voices out there.

I would imagine that ever credit card companies underwritting department does things slighly different, …. I think LC rejects 9 out of 10 applications. I imagine they are not going to release their process about this…(Ive already asked many questions about their process). But when I think about p2p lending, sometimes I say to myself I got to be really stupid for giving someone an unsecured loan to pay off another unsecured loan, without any guarantees that these credit junkies are not going to run up debt again and default. The only thing we can hope for is that they have aggressive collection departments.

BTW: I just seen someone who works for Goldman Sach and other Banks asking for a loan on LC. Now if these guys working for banks and leading investment intutions cannot get a loan or cannot get a competive interest rate through the “normal” way, I dont think I would give them a loan. and how did they pass through the Underwriting process?

Aug. 25, 2011 2:46 pm

@ Peter: 😉

I did my reasearch and do feel p2p is safer and more honest than the stock market.

Aug. 30, 2011 5:48 pm

I’m the creator of Smart Peer Lending. And I thought people might be interested to know that I just posted a full description of the approach and assumptions behind my Loan Analyzer at

I’m using the ROI formula created over at LendStats (thanks Ken!). Primarily because this value is so widely quoted. But thats not where I’ll put a lot of my focus. Over the next few months I’ll be launching features to more deeply analyze loan performance with an emphasis on developing high expectation investment strategies.