|
Chad Sexington posted:We waived inspection and financing contingency and did a partial appraisal contingency to get the house we wanted in the overheated D.C. market. You can still get a pre-inspection before putting in an offer, but there's no question it's a gamble. And I was sweating pretty hard ITT about the appraisal. House buying is like PVP ranking in an mmo. There is an element of pitting your skill against others, but it's more a competition of who is willing to sacrifice more of their time/wealth/health/sanity. The reward is a dubious status symbol, and tokens that the system tells you are great, but are really just more artificial carrots to pull you deeper into the game and further from your humanity.
|
#
?
Feb 15, 2021 21:27
|
|
|
|
| # ? May 15, 2024 13:55 |
|
|
Epitope posted:The reward is a dubious status symbol Sounds like you bought the wrong house. Mine is a home that I enjoy living in.
|
|
#
?
Feb 15, 2021 21:32
|
|
|
Shouldn't have bought that WoW themed house
|
|
#
?
Feb 15, 2021 21:40
|
|
|
The only thing I hope to accomplish (financially) by buying a home is beating the cost of rent over the long term. It sucks rear end that homes are so commodified. I got a contract on a condo where PITI + HOA dues will be less than I'm paying in rent for a smaller apartment in a less convenient area. Looks like I'm going to close next week. SamDabbers fucked around with this message at 22:12 on Feb 15, 2021 |
|
#
?
Feb 15, 2021 22:02
|
|
|
Since there's been some dumping on Redfin as a buyer's agent in the last couple of pages (and maybe more previously, I had like 10k unread posts so just skipped to the last couple of pages), anyone have any thoughts on using them as a seller's agent? Here's our situation: Mother-in-law and father-in-law passed away last year. Wife is an only child and we've got a good handle on all of the legal stuff with the estate. She's inheriting a mostly paid off house. We're still in the process of getting the house cleaned out, and meanwhile we've hired a close family member (who happens to be a licensed general contractor who unexpectedly had some bandwidth) to address some deferred maintenance stuff and handle some long overdue updates - everything is being done properly with appropriate permits/etc. The house is in a hot real estate market, within commuting range of a couple different major metro areas. Like pretty much everywhere else, homes are going under contract with multiple offers in just a few days. We've interviewed several different real estate agents to 1) get a diverse set of opinions on what is worth spending money on and what isn't as we clean the place out; and 2) get a sense for who we're most comfortable with to list the house. One of the realtors we like the best also happens to work for Redfin, who will charge us 1.5% instead of 2.5% as the seller's agent. From my perspective, it seems like we're basically just hiring someone to help us price the place and get it onto MLS/zillow/redfin/trulia/etc., then the hot market will take care of the rest. Am I missing something in Redfin's business model that would make using them as a seller a bad idea? I can see the arguments for how a discount broker would be a liability as a buyer, since a buyer's agent has to hustle so much in this crazy market, but being on the other side of the fence it feels like getting the thing listed as cheaply as possible should really be the priority. NJ Deac fucked around with this message at 23:01 on Feb 15, 2021 |
|
#
?
Feb 15, 2021 22:58
|
|
|
I am not in the market for selling a property, but if I was, I would most likely use Redfin for it. I don't see very many downsides at all plus you get the extra cash. Their marketing is slick but so is their product.
|
|
#
?
Feb 15, 2021 23:02
|
|
|
I get the sense that Redfin attracts a lot of newer agents who are looking to get some experience under their belt and get their name out before moving on to a better fee structure. In a hot market, I think the main job of a good seller's agent would be to 1) maximize your sale price and 2) structure a deal to minimize your risk. If you feel like you need the help of an experienced agent to do that, maybe look elsewhere. If you feel like you understand the selling process pretty well and can do that on your own, and you essentially need a paperwork filler-outer, Redfin might be a good fit.
|
|
#
?
Feb 15, 2021 23:07
|
|
|
NJ Deac posted:Since there's been some dumping on Redfin as a buyer's agent in the last couple of pages (and maybe more previously, I had like 10k unread posts so just skipped to the last couple of pages), anyone have any thoughts on using them as a seller's agent? If you're OK potentially not getting top dollar or the fastest closing for your home, sure go for it. If you just need grandma's house sold they will do that. Make sure you understand a price you're willing to sell it for based on reasonable comps.
|
|
#
?
Feb 16, 2021 01:39
|
|
|
H110Hawk posted:If you're OK potentially not getting top dollar or the fastest closing for your home, sure go for it. If you just need grandma's house sold they will do that. Make sure you understand a price you're willing to sell it for based on reasonable comps. Well, obviously our goal is to maximize the amount of value we get in the sale. If I'm screwing myself by considering a cheaper sellers agent, I want to avoid that - I'm definitely open to listening to why I should go either way. I'm just trying to understand where a sellers agent arguably adds value that I might be making a mistake if I go with redfin. Whether I use a 2.5% agent or a 1.5% agent, aren't they all using showingtime and putting listings on the same websites? Like, we've interviewed 4 agents and they all gave us very similar estimates based on the same comps in the neighborhood, and made similar suggestions for stuff to fix/not fix - one just happens to charge 1.5% instead of 2.5%. How would I potentially not get "top dollar" if I go with the less expensive option?
|
|
#
?
Feb 16, 2021 01:53
|
|
|
NJ Deac posted:Well, obviously our goal is to maximize the amount of value we get in the sale. Then don't hire a budget service provider. You know this is a budget service provider. It's what you said in your initial post about this. There is no free lunch. NJ Deac posted:Like, we've interviewed 4 agents and they all gave us very similar estimates based on the same comps in the neighborhood, and made similar suggestions for stuff to fix/not fix - one just happens to charge 1.5% instead of 2.5%. How would I potentially not get "top dollar" if I go with the less expensive option? Marketing. Like, not just putting it on web sites. This may or may not matter. If you're selling a widget (condo, development house where there are many of the same) then who cares. If you are selling something else you might want to have someone who at least knows the market you are in on your side when advising you about negotiations. This is not a thing you get with redfin, and you have no continuity as your are handed off to multiple agents. How much will you lose? Who knows. If you just want to get rid of the place it it doesn't matter. Motronic fucked around with this message at 02:34 on Feb 16, 2021 |
|
#
?
Feb 16, 2021 02:30
|
|
|
Epitope posted:House buying is like PVP ranking in an mmo. There is an element of pitting your skill against others, but it's more a competition of who is willing to sacrifice more of their time/wealth/health/sanity. The reward is a dubious status symbol, and tokens that the system tells you are great, but are really just more artificial carrots to pull you deeper into the game and further from your humanity. I feel like it's the opposite of a status symbol. It's a private place where my wife and I can do dumb poo poo and enjoy each other and our hobbies without being subject to judgment from the world at large. For sure though ownership gives you more stake in "the game" society has you play.
|
|
#
?
Feb 16, 2021 02:55
|
|
|
Offer #2. Not optimistic about this one, but it's a bit beaten up and the asking price is not as aggressively low as a lot of what we've been seeing so hopefully some people will balk.
|
|
#
?
Feb 16, 2021 03:36
|
|
|
Motronic posted:Then don't hire a budget service provider. You know this is a budget service provider. It's what you said in your initial post about this. The implicit statement you are making here is that the home will sell for more and/or faster by listing with a traditional agent over redfin. Is it not also possible that the home sells for the same amount it would have otherwise and we come out ahead due to much lower fees? In a supply constrained market where homes are selling quickly, I'm skeptical as to how spending more on "marketing" is a sure net benefit. Having met with a few different agents (3 agents charging market rate and 1 from redfin) and having been presented with near identical suggested listing prices and marketing plans, I'm just not seeing where I'm likely to derive additional value from paying an extra percentage point on the seller side. I guess whichever way we go, we can feel good since there's no way to prove we did worse than the alternative, since it's not like it is possible to do a controlled experiment.
|
|
#
?
Feb 16, 2021 03:51
|
|
|
Chad Sexington posted:For sure though ownership gives you more stake in "the game" society has you play. This has hit me haaaaaaaaaaaaaaaard. My wife and I have moved 3x in 10 years for our careers and it's been nothing but good. I'm realizing I'm more risk averse and anxious in my interactions at work because I'm sitting on a $600k illiquid asset instead of a month to month sfh rental and a boatload of free cash flow. I don't like it at all. The only small benefit is I don't have to watch the rental house disintegrate slowly around me from owner neglect -- in other words, I'm free to contemplate how much maintenance costs. Do never buy. Seriously.
|
|
#
?
Feb 16, 2021 03:56
|
|
|
NJ Deac posted:I guess whichever way we go, we can feel good since there's no way to prove we did worse than the alternative, since it's not like it is possible to do a controlled experiment. This is the key. Either way, if you're willing to bet that redfin can do the job for 1% less ($500k sale price that's $5000) then you do you. It sounds like you have a good idea of what the house will go for, so knock yourself out.
|
|
#
?
Feb 16, 2021 04:56
|
|
|
NJ Deac posted:The implicit statement you are making here is that the home will sell for more and/or faster by listing with a traditional agent over redfin. Is it not also possible that the home sells for the same amount it would have otherwise and we come out ahead due to much lower fees? In a supply constrained market where homes are selling quickly, I'm skeptical as to how spending more on "marketing" is a sure net benefit.
|
|
#
?
Feb 16, 2021 14:19
|
|
|
NJ Deac posted:The implicit statement you are making here is that the home will sell for more and/or faster by listing with a traditional agent over redfin. Is it not also possible that the home sells for the same amount it would have otherwise and we come out ahead due to much lower fees? In a supply constrained market where homes are selling quickly, I'm skeptical as to how spending more on "marketing" is a sure net benefit. I will say that plans are very different from execution. We heard very similar-sounding pitches from agents, but the one we went with to sell our condo helped us stage it to the point of bringing her own furniture, actively went and checked out open houses in our neighborhood to see where we were relative to the competition for the purposes of listing price and was proactive about hyping our place to other agents, many of whom were in the same firm as her. Do I know a Redfin agent wouldn't have done those things? As you say, there's no way to run a control, so maybe they would have. But my guess is on the balance we made out. YMMV.
|
|
#
?
Feb 16, 2021 14:30
|
|
|
Chad Sexington posted:I will say that plans are very different from execution. We heard very similar-sounding pitches from agents, but the one we went with to sell our condo helped us stage it to the point of bringing her own furniture, actively went and checked out open houses in our neighborhood to see where we were relative to the competition for the purposes of listing price and was proactive about hyping our place to other agents, many of whom were in the same firm as her. Staging is an excellent point, and the sort of measurable thing that seems like a clear reason to go with a more expensive realtor if it's part of the marketing package. It's also honestly not an area I had fully considered since none of the realtors I've interviewed so far seemed all that interested in doing it. But, maybe I'm just really terrible at picking realtors to interview (we picked 2 that had recently sold homes of similar value in the same neighborhood, one we worked when we sold our condo a decade ago, and Redfin)! What sort of staging is reasonable to expect from a full service realtor? Is that part of what we should expect from the difference in offering 1.5% (redfin) vs. 2.5% (seems to be market in our area) vs. 3% (don't see any homes on the market right now at 3, but maybe that's because they're all selling quickly?)? What does it take to get a home staged super nicely, anyway - do homeowners go out of pocket above the realtor commission? The place will be largely empty by the time we list it, so this seems like an area where there's potentially a measurable benefit from paying a little more money.
|
|
#
?
Feb 16, 2021 15:40
|
|
|
NJ Deac posted:The implicit statement you are making here is that the home will sell for more and/or faster by listing with a traditional agent over redfin. No, that's what you've read into what I said. NJ Deac posted:Is it not also possible that the home sells for the same amount it would have otherwise and we come out ahead due to much lower fees? Anything is possible. NJ Deac posted:In a supply constrained market where homes are selling quickly, I'm skeptical as to how spending more on "marketing" is a sure net benefit. Okay, what about everything else in that paragraph? You chose the one thing you don't think is applicable and skipped the rest of the point. And you're wrong about that: part of marketing is pricing. NJ Deac posted:Having met with a few different agents (3 agents charging market rate and 1 from redfin) and having been presented with near identical suggested listing prices and marketing plans, I'm just not seeing where I'm likely to derive additional value from paying an extra percentage point on the seller side. You're very obviously set on using a budget alternative. You've already talked yourself into this and thought through how to accept any result that occurs.
|
|
#
?
Feb 16, 2021 15:45
|
|
|
NJ Deac posted:
Staging/repairs/cleaning/etc are generally separate from commission. The general item with realtors is that buyers get hosed and sellers get served. That's why buyers don't pay anything for their realtor and you do.
|
|
#
?
Feb 16, 2021 17:25
|
|
|
Motronic posted:No, that's what you've read into what I said. I stated "I'm looking to maximize value from the sale." You stated "Then don't use a budget provider." You are suggesting that, in order to maximize dollars out of the transaction, I am better off using a traditional agent than a cheaper alternative - i.e., that the extra percentage I pay in commission is reasonably likely to result in more dollars out of the home transaction than I'd save from using a cheaper agent. I'm trying to understand why. Motronic posted:Anything is possible. This seems dismissive of the possibility that the home sells quickly either way, and I come out several thousands of dollars ahead since I paid less to my agent. You seem fairly certain that isn't the case, and I'm trying to understand why. Motronic posted:Okay, what about everything else in that paragraph? You chose the one thing you don't think is applicable and skipped the rest of the point. And you're wrong about that: part of marketing is pricing. Here's your paragraph: Motronic posted:Marketing. Like, not just putting it on web sites. This may or may not matter. If you're selling a widget (condo, development house where there are many of the same) then who cares. If you are selling something else you might want to have someone who at least knows the market you are in on your side when advising you about negotiations. This is not a thing you get with redfin, and you have no continuity as your are handed off to multiple agents. You stated "Marketing" and then listed a bunch of reasons why marketing is important (houses are unique, you need someone who knows your unique market). I'm not trying to be selectively responsive here - I don't sell homes all that often, and I don't know what I don't know. That's why I'm asking for advice here. I'm trying to understand what else there is about "Marketing" that I might be missing out on by trying to save money on an agent. From my lay-perspective, it seems like most of the work is arranging to have professional photos taken, writing copy for the MLS listing and websites, and the legwork of getting things uploaded. Someone just mentioned staging. That's an excellent point that I hadn't considered! If having the place professionally staged is something I should expect from a regular agent that I'm not going to get from Redfin, then I'm interested in learning more! I don't know what else there might be, but I'm trying to find out! Motronic posted:You're very obviously set on using a budget alternative. You've already talked yourself into this and thought through how to accept any result that occurs. This is a misunderstanding of my position. My impression from your initial response was that you were strongly in favor NOT using a budget alternative. When you make conclusory statements like "Then don't use a budget provider - there is no free lunch," I want to try to understand why. Maybe this comes across as probing your assumptions in an adversarial way, which isn't my intention. You clearly feel that there's extra value offered by spending more on a real estate agent. Maybe I just picked lovely agents to interview and it's worth meeting with a couple more to fully appreciate where the extra dollars are worthwhile. Thanks for your thoughts.
|
|
#
?
Feb 16, 2021 17:33
|
|
|
NJ Deac posted:The implicit statement you are making here is that the home will sell for more and/or faster by listing with a traditional agent over redfin. Is it not also possible that the home sells for the same amount it would have otherwise and we come out ahead due to much lower fees? In a supply constrained market where homes are selling quickly, I'm skeptical as to how spending more on "marketing" is a sure net benefit. Whether you believe that to be true or not is your decision. If this is the first time you are selling a home, you may want to consider going with a more experienced agent who can better advise you.
|
|
#
?
Feb 16, 2021 17:35
|
|
|
dy. posted:Of course it is possible. If you were to ask a realtor, their proposition to you would be that a more experienced agent is going to know how to maximize the value and minimize the risk of your sale better than a less experienced one. This may include marketing, pricing, negotiation, and advising you on the pros and cons of different offers in a multiple-offer situation. We've sold a couple of homes before, but under very different market conditions. This is our first time ever considering a discount broker, hence all of the questions. I haven't been overly impressed with the level of service I've received from traditional sellers agents over the years (buyers agents have been a different story - clearly having a good buyers agent is a necessity), but like I said above, maybe I'm just terrible at picking realtors to interview for the listing.
|
|
#
?
Feb 16, 2021 18:26
|
|
|
NJ Deac posted:From my lay-perspective, it seems like most of the work is arranging to have professional photos taken, writing copy for the MLS listing and websites, and the legwork of getting things uploaded. Like, this is the bare minimum. A lot of realtors don't meet this bare minimum, which may be part of your confusion. Ignore those realtors completely, and just look at the ones that meet a basic level of competency. Ok, now look at the sales history of those realtors. Some will be closing transactions faster, some will have more of a presence in your local market, some will be representing better houses. You don't have to know why this is the case, you just need to know which realtors are doing better than others, and then try to engage those realtors. There are intangibles to the sales process that cannot easily be measured, so you need to look for proxy values. There is one realtor who does over half the volume in my neighborhood (maybe more than 75%, it's pretty crazy). I would need a strong reason to not use her if I were looking to sell - she's clearly the bottom of the funnel for serious sellers and buyers. As you stated, there is no counterfactual. You'll never know if you get this wrong, so just do whatever feel right. If redfin feels right, do it. NJ Deac posted:We've sold a couple of homes before, but under very different market conditions. This is our first time ever considering a discount broker, hence all of the questions. I haven't been overly impressed with the level of service I've received from traditional sellers agents over the years (buyers agents have been a different story - clearly having a good buyers agent is a necessity), but like I said above, maybe I'm just terrible at picking realtors to interview for the listing. Ya, you might just be bad at it. There is a huge range of quality in realtors and it's pretty obvious, so if you're consistently picking poorly maybe it's time to get some help with that part of the process.
|
|
#
?
Feb 16, 2021 18:36
|
|
|
What are the benefits of a Redfin agent, then? Just the 1% sale difference?
|
|
#
?
Feb 16, 2021 18:47
|
|
|
AmbientParadox posted:What are the benefits of a Redfin agent, then? Just the 1% sale difference? Redfin is trying to commoditize all parts of the house buying and selling process. This is not reasonable in the vast majority of the market but, to paraphrase, if you are buying or selling identical apartment #127 out of 300 that model probably works pretty well on the buying side. Perhaps on the selling side too, but that depends a lot on specifics - consider Tezer's story about one agent having 75% of the business in their neighborhood. Real estate is not a commodity product. It never will be, except for very specific pockets of nearly identical units that have sufficient regular turnover to provide data to very stupid and brittle pricing models like the redfin estimates and "zestimates". So to your question - the "benefits" are a perceived savings on the selling side. I say perceived because not only our entire discussion on "maybe you're not getting top dollar for a property being sold by a group of mostly disinterested early-in-career remote workers" but also because you can actually negotiate that percentage successfully with a traditional agent. On the buy side? I have no idea. I think they say they'll return the rest of their side of the commission. Again, whether having an experienced agent matters or not depends on the home(s) you are looking for and your area. If you're in a place full of identical condos then redfin start to looks like a reasonably efficient home vending machine. Motronic fucked around with this message at 19:09 on Feb 16, 2021 |
|
#
?
Feb 16, 2021 19:06
|
|
|
Motronic posted:Redfin is trying to commoditize all parts of the house buying and selling process. This is not reasonable in the vast majority of the market but, to paraphrase, if you are buying or selling identical apartment #127 out of 300 that model probably works pretty well on the buying side. Perhaps on the selling side too, but that depends a lot on specifics - consider Tezer's story about one agent having 75% of the business in their neighborhood. Why do you think Redfin's/Zillow's price estimate models are 'very stupid and brittle'? I think they have shown, particularly Redfin, their estimates are consistently quite close to the sale price of that property. Not doubting you, just curious on where your opinion is coming from since I like learning about this. And I imagine they put a good amount of investment dollars into innovating and improving these models, since they're important to the industry. https://www.redfin.com/redfin-estimate Separate point but it bugs the hell out of me that realtors can select to hide estimates on certain MLS sites.
|
|
#
?
Feb 16, 2021 19:23
|
|
|
Inner Light posted:Why do you think Redfin's/Zillow's price estimate models are 'very stupid and brittle'? I think they have shown, particularly Redfin, their estimates are consistently quite close to the sale price of that property. Not doubting you, just curious on where your opinion is coming from since I like learning about this. And I imagine they put a good amount of investment dollars into innovating and improving these models, since they're important to the industry. Both of their pricing models are total crapshoots in my market outside of condo/townhome developments and other fairly new tract housing. Why is it brittle? Because it can't intelligently make decisions about comps. The data to automate this simply doesn't exist in all markets. Many relevant property specifics are simply missing. Also, without sufficient turnover it's left without any sort of intelligence from experience to be able to accurately extrapolate. I don't need to read a buzzword filled article on how they do this: first hand experience in my market demonstrates it doesn't work. How many other markets does it fall on its face in? I don't know, and they're not gonna tell you. But even me, a complete amateur, can and have called sale prices consistently more accurately than either of those service's estimates on homes I had looked at the last time I was in the market to buy.
|
|
#
?
Feb 16, 2021 19:27
|
|
|
Zillow also uses automated data gathering and consequently I've seen many times where it has multiple facts completely wrong. It thought my old single family attached house was a condo (??) with one less bedroom and bathroom than it has and its zestimate was like half of what it should be lol. Meanwhile identical houses (like, clearly same time, same builder) down the block in worse or in one case better shape than ours had zestimates all over the place but in reality were all worth just a bit less than ours or a bit more. Depending on the market, zillow can be truly terrible. The one thing it's nice for is easily viewing listings on a map. That's about it.
|
|
#
?
Feb 16, 2021 19:33
|
|
|
alnilam posted:Zillow also uses automated data gathering and consequently I've seen many times where it has multiple facts completely wrong Yeah, I should have been more clear about that. When I say the data doesn't exist I meant accurately/properly in property records they are using. They try to automagically backfill this data from other sources as well as by begging the owners to submit information. When this data is now part of their property record and is incorrect you ends up with hilariously bad results. Even with accurate data the results are very often terrible.
|
|
#
?
Feb 16, 2021 19:35
|
|
|
Inner Light posted:Why do you think Redfin's/Zillow's price estimate models are 'very stupid and brittle'? I think they have shown, particularly Redfin, their estimates are consistently quite close to the sale price of that property. Not doubting you, just curious on where your opinion is coming from since I like learning about this. And I imagine they put a good amount of investment dollars into innovating and improving these models, since they're important to the industry. Redfin says their estimates are quite close. Redfin says they put a good amount of investment dollars into innovating. Redfin says they are important to the industry. Why do I trust what Redfin says about their own product?
|
|
#
?
Feb 16, 2021 19:39
|
|
|
Inner Light posted:Why do you think Redfin's/Zillow's price estimate models are 'very stupid and brittle'? I think they have shown, particularly Redfin, their estimates are consistently quite close to the sale price of that property. Not doubting you, just curious on where your opinion is coming from since I like learning about this. Well I just looked up our unit. There's more than 300 units in our building, basically two floor plans, 1 bedroom/studio and 2 bedroom, I think 20 units have turned over since we bought ours, redfin and Zillow have a $200,000 spread between them. We're in a key market for their algorithms and there's lots of comparable units in our neighborhood as well. Are they off by more than 30%? No. Could a trained blind monkey hit inside that range? If it were a well trained blind monkey They're good for deciding, "well, how much is a comparable home in long island? Malibu? Hawaii? This is slightly better than a wild guess and I've done no other research so let's give this thing a spin to get started" also good for determining general price trends for neighborhoods etc. Should you use it to make an offer? Probably look at the "sold" tab, not "for sale" Hadlock fucked around with this message at 19:47 on Feb 16, 2021 |
|
#
?
Feb 16, 2021 19:43
|
|
|
alnilam posted:Meanwhile identical houses (like, clearly same time, same builder) down the block in worse or in one case better shape than ours had zestimates all over the place but in reality were all worth just a bit less than ours or a bit more.
|
|
#
?
Feb 16, 2021 19:45
|
|
|
NJ Deac posted:Staging is an excellent point, and the sort of measurable thing that seems like a clear reason to go with a more expensive realtor if it's part of the marketing package. It's also honestly not an area I had fully considered since none of the realtors I've interviewed so far seemed all that interested in doing it. But, maybe I'm just really terrible at picking realtors to interview (we picked 2 that had recently sold homes of similar value in the same neighborhood, one we worked when we sold our condo a decade ago, and Redfin)! Totally depends on the agent. Some will just refer you to a stager that will charge a few thousand per month. Some will offer to do it themselves. Our agent went room by room and told us what to put in storage and what to keep, then went and bought some things to tie the rooms together and filled gaps with her own furniture. A very valuable service that she didn't charge us anything for, but we negotiated it as part of her commission up front. (This is actually a surprisingly touchy area if you have a spouse like mine who likes the way they have furnished your home. A staged home is an exercise in deception, making a room seem bigger than it is by keeping it as empty as possible while also seeming furnished. A good furnished home != a good staged home.)
|
|
#
?
Feb 16, 2021 19:46
|
|
|
Huh I should check to see how well Redfin does estimates on my block. Within about a year, one house sold for $500k and another one sold for $50k. The condition of the houses can be anywhere between exquisite high-end restoration and dilapidated shitheap, but they are otherwise architecturally and dimensionally similar.
|
|
#
?
Feb 16, 2021 19:50
|
|
|
When you learn neural network computing, house pricing is a classic dataset to work with as a lesson. That's because the input is a bunch of facts and values like sqft, lot size, year built, neighborhood, lat/long, # bedrooms, ocean view yes/no, etc, that obviously together have an impact on a numeric output (price) but in a non obvious way. This is exactly the kind of problem neural networks excel at, it's almost the perfect example dataset to explain the whole approach to machine learning. I suspect someone took machine learning 101 and said hey wow maybe I could make a business out of that? and founded zillow and redfin. The problem is there are a lot of reasons it's not actually a great way to price any one individual house. In fact some of those reasons are also classic "shortcomings and limitations of machine learning" issues, which plays even more into making it a good teaching example, and is the kind of thing that person should have learned in machine learning 102 but possibly didn't. For one, there is a lack of data. This kind of model works best with like, thousands or millions of data points to train from. The model really needs to be regionally distinct for obvious reasons, and in any given city there is likely not enough movement to train a great model, not to mention the market changing over time so you have either de-weight or even throw out older data all the time. There are tons of uncaptured variables whose quantification and collection can be difficult or impossible but are extremely important, like state of repair, state of update, is it on a busy street, is it next to something loud or unpleasant, etc. Then there's the obvious fact that house buying is very subjective, like even if you did somehow quantify and gather data on state of update, what if someone with terrible taste did the updating. What if it's listed as "ocean view" but the ocean view is lovely (there lies another fun issue is if you rely on self-reporting for a lot of this stuff, people will tend to exaggerate to game the system and claim an ocean view because you can kind of see a glint of blue between two houses and through a tree if it's winter and the leaves are gone). These models probably work okay in aggregate. Meaning if you are a real estate investor with millions of dollars, and you trust this model to make decisions on hundreds of properties, you will probably end up benefiting from the guidance of this model, because it will be close enough a lot of the time and its under-estimates will probably balance out with its over-estimates. But if you are an individual homebuyer, it means jack poo poo. And by the way, guess where Redfin is pivoting - that's right, large-scale real estate investment.
|
|
#
?
Feb 16, 2021 19:58
|
|
|
Has anyone done a USDA farm loan Looks like you already need to own a micro farm, or have managed someone else's farm for at least three years to get one Kind of looking at buying an acre of both rural and wildfire scarred land that's cratered in value, and then digging a well + planting an orchard (almonds are popular, but extremely destructive to the local water table, so probably not those) Looks like you could buy an acre of mostly buildable vacant land, dig a well and plant ~75-150 walnut trees per acre. That would probably qualify you for a larger farm loan after three years. Land plus well plus trees minus labor probably fits inside the magic $55,000 tax write off/small business pass through amortized over six years thing that passed not too long ago
|
|
#
?
Feb 16, 2021 20:06
|
|
|
alnilam posted:When you learn neural network computing, house pricing is a classic dataset to work with as a lesson. That's because the input is a bunch of facts and values like sqft, lot size, year built, neighborhood, lat/long, # bedrooms, ocean view yes/no, etc, that obviously together have an impact on a numeric output (price) but in a non obvious way. This is exactly the kind of problem neural networks excel at, it's almost the perfect example dataset to explain the whole approach to machine learning. Zillow funded a $1.2m public competition on this very issue, and you can see all sorts of discussion of the problem and accuracy analysis here: https://www.kaggle.com/c/zillow-prize-1/overview Indeed, the models work very well on large-scale datasets. But, of course, any specific house or street may have quirks (or data inaccuracies) that make market value depart from ML estimate value. Paired with human agents who can adjust for those problems, getting directly into the real estate market is an intriguing business by redfin / zillow.
|
|
#
?
Feb 16, 2021 20:08
|
|
|
alnilam posted:When you learn neural network computing, house pricing is a classic dataset to work with as a lesson. That's because the input is a bunch of facts and values like sqft, lot size, year built, neighborhood, lat/long, # bedrooms, ocean view yes/no, etc, that obviously together have an impact on a numeric output (price) but in a non obvious way. This is exactly the kind of problem neural networks excel at, it's almost the perfect example dataset to explain the whole approach to machine learning. To flip it around, if you built a model of home prices and use that to assess risks in lending, you'll wind up disproportionately harming minorities because your model says their homes are worth less on average.
|
|
#
?
Feb 16, 2021 20:36
|
|
|
|
| # ? May 15, 2024 13:55 |
|
|
Motronic posted:This is not reasonable in the vast majority of the market but, to paraphrase, if you are buying or selling identical apartment #127 out of 300 that model probably works pretty well on the buying side. What's the goon census on buying a generic apartment in a typical default big city high rise? Beyond normal buying advice such as deferred maintenance or HOAs is there anything else I should be looking at? The context behind this is I'm originally from Southern California but it's ridiculous expensive for my budget but I could eventually save up. I'm thinking that I could buy a smaller place in a more affordable part of the Country for a few years and with my equity use that to get something back in Los Angeles. Along with the hope that the housing crisis improves as well as it couldn't possible get any worse? 
|
|
#
?
Feb 16, 2021 20:36
|
|
