Memoirs of a Money Manager: Plumbing the Data: How Correlation Can Lead and Mislead

Many unusual events took place during our long stay, due to both the questionable but rapidly improving neighborhood and the requirements New York City placed on townhouse owners. In general, because the renovation left the building in a “like new” condition, we had few problems with the building’s functions. One problem we did have, however, was an occasional flooded basement.

As many property owners know, leaking roofs, leaking windows and plugged toilets go with the territory, but our basement flooding was strange. While occasional flooding after prolonged spring rains can be attributed to overflowing city storm drains, our problem was different. Our basement flooded in the winter, but not every winter, and not in any easily identifiable pattern. It was so intermittent that I would forget about it until a few years later when it would flood again.

One frigid January Friday, I needed some old files from the basement storage room. It was about 5:30 p.m., all the employees had left earlier due to the bad weather, and I had ordered in Chinese food for dinner. I raced down the steps eager to find the files and get back upstairs before my food arrived. Splash, kaboom! I hit the wet floor, went flying and narrowly missed doing serious damage to body and soul. Three or four inches of water covered the entire basement. I waded into the storage room, retrieved the file and returned to my office just in time to fetch my dinner.

As I supped on my beef and broccoli, brown rice and hot tea, I pondered the problem of the flooded basement. It was sporadic and caused no real harm because we had built the storage shelves one foot off the floor. I had not thought much about it, but this was our third flood in the last week, so I was concerned about the trend. Unfortunately, I could not spot a trend. I had to work late that night and all weekend on a special project for a client so, when at about 9 p.m. I went to visit the basement bathroom at the foot of the stairs, I was puzzled to find the basement dry. I decided to check it out more frequently; I also decided to spend the night in a small apartment I had in the back of the basement since it was cold outside and getting a taxi home would be difficult.

At the foot of the basement stairs in the front of the building was a half bath; the storage room occupied the third of the basement closest to the street. The oil furnace claimed the middle third and the back third was my “apartment.” To call it an apartment was admittedly a stretch. I had furnished it with a closet where I kept a few white shirts, ties and a couple suits in case a client made an unannounced visit, a full bath with shower, a dry heat sauna in which I spent many wonderful winter evenings contemplating life, and a single bed where I often spent the night if I had to get up very early the next morning or I was working over the weekend. The entire space was elevated about five or six inches above the main basement floor and virtually never flooded.

Sometime after 10 p.m. I went downstairs to check the basement and was shocked to see that an inch of water covered the floor; I went upstairs to retrieve a ruler to measure the depth. Over the next hour the water depth steadily increased, peaking at slightly over 4 inches at 11:30 p.m. and then receding to just a damp floor by about 12:30 a.m. As I sat there measuring the water it suddenly hit me: the water was warm! Thinking about it, I also recalled that all the time I was running up and down the stairs to measure the water the furnace was huffing and puffing; it was silent now and had been for a while. Of course, there must be a leak in either the pipes going to the steam radiators upstairs or in the return lines. Problem solved. I would get the plumber in here on Monday and get this problem fixed once and for all. I spent the night in the basement apartment and on Saturday morning awoke to the clanking pipes as the furnace starting heating up the building. All day the furnace cycled on and off and the warm water level rose and fell in perfect correlated rhythm. I finished my project early Saturday evening and went home to clean up, meet some friends and have some real food. Sunday warmed up and I went to the movies. I was self-satisfied having cracked the mystery of the flooding basement.

The plumbers—they always come in pairs—came on Monday and, of course, the basement was not flooded due to the warming spell. So, they cranked up the boiler, found a small leak in a return pipe, replaced it and wrote up their outrageous bill.

The warming trend stayed in place for some time, winter departed, spring came and that was that.

But unfortunately that was not that. My subconscious kept hammering at me about the darn flooding; something did not make sense. The flooding never was a serious problem, nor did I expect it would be in the future, I just thought that I had missed something. But the next winter came and went, it was mild and the basement did not flood. I forgot about it.

A couple years after the plumber “fixed” the problem, I was on vacation in the Caribbean and my office manager called me to report that the basement was flooding. She had called the plumber and he claimed that the boiler must have a fine crack and would have to be replaced. It was not going to be cheap. My instinct was to just have it done, but the weather was really cold and the office would be freezing for a couple of days without any heat; of course they could do the work on the weekend at time-and-a-half or double time. I decided to hold off until I got back and the weather warmed.

I returned on a Saturday, but not only had the weather not warmed, it was now in the low teens and blusteringly cold. My office manager had left a message for me that the basement had been flooded off and on all week. I rose early Sunday morning to visit the scene of the crime. I arrived at the office fortified with bagel, lox, cream cheese, coffee and the Sunday New York Times. The furnace was huffing and puffing and the old radiators were clanging as they do; the basement was at low tide, but rose steadily over the next couple of hours. When the building was warm the huffing and puffing waned and I immediately went to the basement to measure the high tide level; just under five inches. Slowly, but steadily, over the next couple of hours, the water drained away and the basement became, if not dry, at least without a noticeable ocean. I resolved to once and for all find where the leak was in my furnace.

The plumbers had reported that the real problem was a crack in the boiler. If so, then this problem was a new one, unrelated to the leaking pipes that the plumbers had fixed a couple of years earlier. I decided that as soon as the huffing and puffing started I would take a folding chair to the furnace room and plant myself under the sole hanging light bulb and, with the Times in hand, wait and watch. Late in the afternoon the outside temperature began to drop and the outside “heat timer” cued the furnace to send heat. Down to the basement I went, coffee and paper at the ready. Planting myself directly in front of the furnace I read and waited; I knew from previous measurements that it took quite a while for the water level to get up to a measurable level but I was determined to see the leak first hand. The first hour went by and nothing happened; no water around the furnace. After about an hour and half I rose to go to the bathroom and, pushing the heavy steel fireproof furnace room door open I was met with a “tidal wave.” Well, not really, but the hallway was damp and starting to flood. Clearly, the plumber’s “cracked boiler” theory did not hold water, so to speak. It was now close to 7 p.m. and I was tired, so I decided to get some takeout Indian food and spend the night in my little apartment and “sleep” on the flooding problem.

I called my office manager to tell her of the problem and that I would be in the building when she or the receptionist arrived, but the alarm might be set. Out I went into the night, and it was indeed cold and windy. My favorite Indian takeout place was open and I decided to eat there and enjoy the food and the crowd.

This respite got me thinking and gave me an idea about the flooding. The facts were puzzling. I retrieved a pen from my pocket, grabbed a napkin and began to make notes. First, it did not flood every year. This meant it was different problems in different years as the plumbers claimed or, that the problem of flooding was related to something else that they had not yet identified. Second, it only flooded after cold weather that persisted for several days and these conditions did not occur every year. Third, the flooding cycled on and off in perfect lagged response to my furnace responding to the call for heat from the outside heat timer thermostat. Fourth, it most decidedly did not come from the furnace boiler or from any pipes inside the boiler room because I had watched the boiler and the furnace room was dry when the outside hallway was starting to get damp. Fifth, the water was warm, not hot, so maybe it came from the return pipes that carried it back to the boiler. Sixth, since the furnace was constantly losing water, this implied that the boiler must be taking on new water to replace the water lost through the leak. Lastly, I had discovered that afternoon that my furnace outside heat timer was only a few feet from my next door neighbor’s heat timer; thus, it was very likely that both furnaces would activate at the same time.

These were the thoughts I went to sleep with that night. My apartment was right next to the furnace room so I could hear the huffing and puffing and clanging of the pipes and the radiators all night. I was exhausted; sleep came easily and certainly by 9 p.m. I was gone to world, but my brain just would not leave it alone.

About midnight I awoke with a great insight, or so I thought. I rose quickly, for if I was to test my theory I needed to do it before my employees arrived in the morning. The furnace was quiet, the building was warm and the water was receding and barely an inch deep. At the top of the stairs there was a big red switch plate that acted as an emergency turn off switch for the furnace; I turned it off, set an alarm for 4 a.m. and went back to sleep.

When the alarm sounded, I leaped out of bed and went straight to the hall outside the furnace room. The building was cold, but I was met by several inches of warm water! There was no doubt it was a furnace leaking—just not mine. I went upstairs, flipped on the switch to start warming up my building and went back to sleep. I slept very well.

The townhouse next to me and adjacent to the hallway that flooded was owned by a somewhat cranky old lady who had occupied it since the late 1940s. Over the years we had developed a rough but workable relationship. She was well educated, highly intelligent and well-traveled, having been the editor of a food and travel magazine. The next morning, when my basement was at high tide, I called her.

“Gerry,” I said. “Can you please check your basement and see if you have a leak from your furnace that is flooding your floor; the water should be a little warm. If you could please check now I would appreciate it. My basement is flooded and I think it is coming from under your wall adjacent to my wall.”

“Sorry,” she said. “My basement is dry. I suggest you call your plumbers.”

“But, Gerry, I—“ the line went dead as she hung up in my ear.

What to do? I was absolutely convinced that the problem was coming from her furnace. Nevertheless, I decided to pay a visit to the townhouse next to Gerry’s, or one building away from mine. I did so within minutes of my call to Gerry and, after explaining my problem, was permitted to look at their furnace room. Their furnace was cranking out heat, but the floor was dry. The owners were very good friends of Gerry, but also sympathetic to my dilemma.

They asked a question of me, which ended up providing a solution to the problem. “Have you made an insurance claim yet?”

“No,” I replied, “Because I have no damage.”

Their question gave me an idea. I hurried to the building on the other side of me; it was an apartment building with a super. He was very accommodating and took me to their furnace room. Dry.

I called my insurance agent, explained the problem. I was insured through Chubb and they agreed to send a claims adjuster the next day. I explained that the forecast was for a warming trend to start in a couple days and I needed cold weather to prove my case.

The next day the insurance adjuster came and I showed him how the basement flooded from under the wall from Gerry’s building when my furnace was off. He requested I call Gerry again and, with him on the phone, she once again confirmed her furnace was running but that the basement was dry. I explained to the adjuster how I had visited all three neighbors and he proceeded to do likewise. My insurance claim, I asserted, was that with continually flooding of water under the wall my building foundation and wall were at risk of structural damage. The adjuster agreed. He promised to be back to me within a few days; actually he called back the following morning. As luck would have it, Gerry was also insured through Chubb and they arranged for a plumbing firm to visit both our houses at the same time.

The plumbers came later that morning and, thankfully, it was still cold. Because Gerry’s basement was dry they clearly did not believe me, but I suggested that they could see that with my furnace off and Gerry’s furnace running the water would flow right under the adjacent wall. In addition, I suggested two other tests. First, they could shut down Gerry’s furnace and run mine and see if there was any flooding in my basement. Second, they could put dye in Gerry’s furnace water and see if the water that came into my basement was colored. In the end they did all of the above and, of course, the colored water flooded under the adjacent wall. The great unsolved puzzle was that Gerry’s basement was dry throughout all these tests.

As it turned out, in the early 1950s Gerry had her furnace replaced. At that time all the return pipes from her building ran across the basement floor. Rather than remove them, they simply cut them off, poured a new concrete floor and installed the new furnace on the new floor; except they missed a few return pipes buried in the walls. Only if the weather remained cold for several days, requiring nearly continuous running of the furnace, did enough pressure build up in the return pipes to force water through the pipes. Over many years, the water seeped through the old floor and into my basement, which was a few inches lower. Of course, there were many winters where that condition was not met. In any event, that is how the insurance adjuster explained it to me.

Gerry was initially furious with me for bringing in Chubb, but when she realized that the plumbers would be ripping her furnace room apart looking for the offending pipes, she was delighted that the claim would cover the work. I had no claim except a minor amount for a structural engineer to examine her wall and my wall to confirm that there was no long-term damage.

A few weeks later Gerry bought me lunch at a nearby restaurant; our friendship warmed, as did the weather. I had learned another lesson in problem solving. Because Gerry’s furnace outside heat timer and mine were in close proximity they called for heat at about the same time. This meant both furnaces were correlated to a third variable, the outside temperature. Only by turning off my furnace and letting hers run could I prove that the problem was not my furnace. The fact that her basement was dry was a misleading distraction that confused the issue. After the fact it all seemed so obvious. But then it always does.