OPINION
HELEN G. BERRIGAN, District Judge.
This matter concerns the alleged malfunction of an autopilot system onboard a tug, the MTV BLUEFIN, leading to a collision between the BLUEFIN’s barge, PENN 80, and another barge, the OSG 400, in the Delaware River on November 5, 2010. The claims and defenses in this matter were tried to the Court with no jury on May 19 and 20, 2014. Having considered the testimony of the witnesses and all of the evidence, the Court determines that plaintiff Penn Maritime, Inc. (hereinafter “Penn”) has failed to meet its burden of proof that this incident was caused by any defect in the BLUEFIN’s autopilot. As explained below, it is at least equally probable that this accident was caused by the failure of Penn’s captain to properly operate the autopilot. Accordingly, the Court awards judgment in favor of defendant Rhodes Electronic Services Inc. (hereinafter “Rhodes”) on Penn’s main demand and in favor of third-party defendants, Thoma-Sea Marine Constructors, LLC and Thomassie Properties LLC, f/k/a Thoma-Sea Shipbuilders LLC (hereinafter “Thoma-Sea”) on Rhodes’s maritime inter-pleader.
STATEMENT OF FACTS
The M/V BLUEFIN is an Articulated Tug Barge (ATB) owned and operated by Penn.1 As an ATB, the BLUEFIN can either push or pull a barge.2 In push-mode, it connects to the rear end of a barge via retractable pins in its bow.3
Penn contracted with Thoma-Sea to build the BLUEFIN.4 A company named Engine Motor, Inc. (hereinafter “EMI”) installed the vessel’s basic steering system.5 The EMI steering system includes, among other things, a lever for hand steering, and a console to switch between hand steering and autopilot, assuming that an autopilot system is installed.6 The EMI system is not equipped with its own autopilot.
Penn hired Rhodes to install the BLUE-FIN’s remaining electronic navigation equipment, including a Simrad AP50 Plus Autopilot system, which Rhodes did after EMI had already installed the steering.7 The initial installation was performed by multiple technicians from Rhodes, including Branden Rhodes.8 After the installation, Rhodes performed a full “dockside setup,” giving the autopilot some basic in[511]*511formation about the vessel, such as length, and the preferred settings for operation.9 Rhodes personnel also instructed the Penn crew in the basic use of the autopilot.10
At the end of July 2010, the BLUE-FIN participated in sea trials to test its critical equipment in realistic operating conditions.11 The sea trials were conducted without the barge PENN 80, which was also a recent construction.12 By all accounts, the autopilot functioned well during sea trials.13 However, the crew reported that hand steering was slightly misaligned to the starboard side of the vessel.14 EMI corrected this issue to Penn’s satisfaction.15 Around this time, Captain Thornton also complained that the BLUEFIN had a high level of critical vibration relative to his previous vessel.16 Penn did not opt to have this issue looked into or repaired.17 Penn took delivery of the BLUEFIN after sea trials, certifying that all systems on board were in working order.18
From September 4 to 6, 2010, the BLUEFIN took its maiden voyage with the PENN 80 from New Orleans, Louisiana to Mobile, Alabama.19 En route, Penn’s crew experienced a technical difficulty with the operation of the autopilot. The crew members differed in their explanations of the problem. Captain Thornton testified that the BLUEFIN would not hold its course from the very beginning of the voyage and navigated in a series of “S” loops, even before it took on any cargo.20 Mate Washburn testified that the autopilot struggled to maintain course, but only after the barge was laden with cargo and, specifically, loaded down at its head.21 Whatever the issue, Captain Thornton attempted to change settings on the autopilot in response using the installation manual.22 The end result was that the autopilot had an error message and would not function at all.23 Captain Thornton hand steered the BLUEFIN through the rest of the maiden voyage, and reported “problems” with the autopilot to Penn’s representative, Dan Duplantis.24 He also reported that the BLUEFIN had regained the “starboard draw” in hand steering that it had during sea trials.25
Duplantis called EMI to fix the hand-steering issue.26 On September 15, 2010, Branden Rhodes came on board the BLUEFIN for the express and limited purpose of correcting the error message on the autopilot.27 He changed the autopi[512]*512lot’s drive setting from solenoid to analog.28 He also recalibrated the analog drives.29 The analog drives in the autopilot control the rudders through the main steering system using electrical voltage.30 The drives have to be calibrated in order to set an amount of rudder movement for a particular voltage.31 Branden Rhodes had no recollection of altering any other settings, like vessel length, trim mode, and sea state.32 He did not inspect the junction boxes below the bridge where the autopilot was wired into the rest of the steering and navigation equipment.33
Following this repair, the BLUEFIN reentered the Rhodes fleet under the command of Captain Eric Edvardsen.34 Captain Edvardsen had been present for the installation of the steering and navigation systems and the sea trials.35 He had some familiarity with the autopilot system from his previous experience with a similar model onboard his previous vessel.36 According to the deck logs, Captain Edvardsen had command of the BLUEFIN from September 15, 2010, when Rhodes did its reconfiguration of the autopilot, until October 7, 2010.37 During that time he used the autopilot without incident.38 On October 7, Captain Thornton relieved Captain Edvardsen.39 During his 21 days on the vessel, he also operated the autopilot without incident.40
Captain Edvardsen rejoined the BLUE-FIN on October 27, 2010 in Houston, Texas.41 The BLUEFIN was preparing to tug a full load of No. 6 fuel oil in the PENN 80 barge from Houston to Paulsboro, New Jersey.42 During this trip, and prior to the events at issue in this case, he operated the autopilot numerous times without any problem43
On November 5, 2010, when the BLUE-FIN reached the Delaware River, Captain Edvardsen went to relieve the mate on watch, Mate Washburn, who did not have Delaware River pilotage approval.44 In order to obtain that approval, Mate Wash-burn needed 12 supervised round trips on the Delaware.45 Thus, Mate Washburn remained in the bridge to observe Captain Edvardsen after he was relieved.46
For navigation purposes, the Delaware River is divided into various ranges, begin-[513]*513rang with Brandywine.47 Captain Edvardsen navigated the first four of these ranges on autopilot without changing to hand steering, making slight changes to the BLUEFIN’s heading using the autopilot’s course adjustment knob.48 He planned to change to hand steering to make a significant turn between the fourth and fifth ranges — Liston and Baker, respectively.49 Just before the turn, the Captain saw an oncoming vessel, the OSG • CONSTITUTION, on the opposite side of the channel.50 The CONSTITUTION was tugging a barge of its own, the OSG 400.51 Both the BLUEFIN and the CONSTITUTION were traveling at roughly 10 knots.52
At the turn, Captain Edvardsen changed over to hand steering as planned, with Mate Washburn watching from behind.53 In the Liston Range, the BLUEFIN had been tracking a 317 degree heading.54 Captain Edvardsen made the turn and then brought the BLUEFIN to a 359 degree heading to bring it to the right side of the channel.55 He steadied up the vessel, making sure that the heading on the gyro-compass matched the heading on the autopilot console.56 Then, Captain Edvardsen restored control of the vessel to the autopilot.57 The BLUEFIN immediately “sheered.” Its rudders went “hard over” or as far as they could go, causing the vessel to turn as sharply as possible to port.58 By the time Captain Edvardsen and Mate Washburn realized what had happened, the BLUEFIN had already crossed the channel on a collision course with the CONSTITUTION and its barge.59 Captain Edvardsen stopped the engines and tried to hand steer to avoid the collision.60 The BLUEFIN eventually stopped but not before the PENN 80 dealt a glancing blow to the OSG 400.61 Fortunately for all, no oil spilled.62
After the collision, Captain Edvardsen radioed the Coast Guard, who told him to go to the nearest anchorage, an artificial island in the Delaware.63 The Coast Guard tested the Captain and Mate Wash-burn for drugs and alcohol.64 The results were negative.65 Afterward, the Coast Guard asked the Captain to replicate the sequence that resulted in the accident.66 Captain Edvardsen did so, transferring from hand steering to autopilot as he had done after his turn, and just as in the accident, the rudders went hard over to port.67 He repeated this process twice [514]*514more for the Coast Guard, and twice more, the rudders went hard over to port.68
The following day, on November 6, 2010, Captain Edvardsen repeated the process for an inspector from the American Bureau of Shipping, while the BLUEFIN was docked at a terminal in the Port of Wilmington, with the same results.69 Later that day, Penn telephoned Tom Pisciotta, a local Simrad-certified dealer and repairman, to repair the autopilot.70 Pisciotta came immediately.71 He went with Captain Edvardsen to the bridge where the Captain explained that the BLUEFIN was inexplicably sheering.72 Pisciotta then attempted to test this issue by switching from hand steering to autopilot and observing the reaction of the rudders.73 He performed this test multiple times, each time allowing the autopilot heading to match the gyrocompass before engaging the autopilot.74 Each time, instead of remaining at midships, the rudders went over to port by between 8 and 10 degrees.75 On one occasion, the rudders went slightly to port when Pisciotta attempted to make a starboard turn using the autopilot controls.76 At no point was he able to get the boat to sheer as it did during the accident.77
Believing that there was a malfunction, Pisciotta went below the bridge to examine the junction boxes that contain the wiring and circuitry for the autopilot.78 He found that some of the screws that hold the wiring in place were loose.79 He also found that a spare cable was pressing against the center of an analog board.80 Pisciotta tightened the screws from between three-quarters and one full turn and removed the spare cable.81
Next, Pisciotta went back to the bridge and changed certain settings on the autopilot console.82 He changed the vessel length to accommodate the BLUEFIN and PENN 80 barge together, whereas it had previously been set to a length that would only accommodate the BLUEFIN alone.83 He turned off the Auto Sea State and Auto Trim settings.84 He recalibrated the rudder end points, which had been set well below manufacturer recommendations.85 Finally, Pisciotta changed the rudder zero point to zero from around 8 degrees.86 After all of these changes, the autopilot performed normally.87
PROCEDURAL HISTORY
Penn brought suit against Rhodes in November 2011 for negligent installation [515]*515and maintenance of the Simrad AP50 Plus autopilot, breach of contract and warranty, and products liability.88 Penn simultaneously sued Navieo, the autopilot’s manufacturer, for products liability and breach of contract and warranty.89
On February 24, 2011, Penn amended its complaint to add the owners and operators of the OSG CONSTITUTION and the OSG 400 as defendants, alleging that negligence in the operation of those vessels had caused the accident.90 In June 2012, Penn filed a second amended complaint to substitute the OSG CONSTITUTION/OSG 400 defendants.91 In answer to this second amended complaint, Rhodes filed a third-party complaint against Thoma-Sea under Fed.R.Civ.P. 14(c), alleging that excessive vibrations inherent in the BLUE-FIN’s construction were responsible for the autopilot malfunctioning at the time of the accident.92 Following notice of bankruptcy, Penn voluntarily dismissed all allegations of fault pertaining to the owners of the OSG CONSTITUTION and OSG 400.93
Navieo obtained summary judgment on all claims against it in April 2014.94 The Court entered partial final judgment as to Navieo, as well as other entities, by agreement of the parties, in a judgment dated May 1, 2014.95 Penn’s claims of negligence, breach of contract/warranty, and products liability against Rhodes and Rhodes’s interpleader against Thoma-Sea were tried to the Court on May 19 and 20, 2014 and taken under submission thereafter.96
LAW & ANALYSIS
Penn has raised three claims against Rhodes: negligence, breach of implied warranty of workmanlike performance (WWLP), and products liability.97 Penn claims that Rhodes committed negligence and breached the WWLP by improperly installing and setting up the Simrad AP50 Plus autopilot system on the BLUEFIN in July and August 2010 and/or failing to make proper repairs on September 15, 2010 when a malfunction was reported to them. They cite the specific issues noted and changed by Tom Pisciotta when he serviced the autopilot on November 6, 2010: (1) a spare cable pressing on analog motherboard in the JD53; (2) loose wire screws in the JD53; (3) Auto Sea State improperly engaged; (4) Auto Trim State improperly engaged; (5) improperly set boat length; and (6) improperly calibrated rudder end points and zero point.
I. Ordinary Negligence and Breach of WWLP
A. Legal Standards
“To establish maritime negligence, a plaintiff must ‘demonstrate that there was a duty owed by the defendant to the plaintiff, breach of that duty, injury sustained by [the] plaintiff, and a causal connection between the defendant’s conduct and the plaintiffs injury.’ ”98 “Wheth[516]*516er a defendant owes a plaintiff a legal duty is a question of law.”99 A duty is owed only with respect to an interest that is foreseeably jeopardized by an act or omission.100 A harm is not foreseeable unless it “might have been anticipated by a reasonably probable result of the act or omission ....”101 A marine service contractor has a duty to exercise reasonable skill and care 'commensurate with the knowledge normally possessed by members of his profession.102
“Those who repair a vessel or the equipment aboard it make a warranty, the implied warranty of workmanlike performance.” 103 To have a cause of action for breach of implied warranty of workmanlike performance, a shipowner must show that the contractor breached the warranty and that this proximately caused the shipowner’s injury.104
The implied warranty of workmanlike performance is breached where a contractor fails to perform his obligation properly and safely.105 However, where the defendant properly performs the “essence” or “inescapable elements” of a contract, no cause of action for implied breach will lie.106 In practice, this standard need not differ from the test for ordinary maritime negligence.107 An obligor generally complies with the WWLP when he uses reasonable care in the performance of the obligation.108 The requirements of causation and foreseeability still apply to claims under the WWLP.109
Penn has the burden of proof on all of its claims by a preponderance of the evidence.110 “A preponderance of the evidence simply means evidence that persuades [the Court] that the plaintiffs claim is more likely true than not true.”111 Circumstantial evidence may help the plaintiff meet this burden.112 However, where only circumstances are relied upon, they must permit a strong inference on the required [517]*517elements of the plaintiffs claim.113 In other words, the circumstances must exclude other reasonable hypotheses with a fair degree of certainty, showing that the defendant’s liability for negligence is more likely than not.114 In marine repair or servicing cases, the identity of the party in control or possession of the vessel at the time of the incident is a factor to consider when determining whether circumstantial evidence of negligence or causation is sufficient.115
B. Analysis
1. Penn has not proven Rhodes’s responsibility for the allegedly “improper” autopilot console settings identified on November 6.
As an initial matter, Penn has alleged that this accident was caused by a number of settings (analog-drive calibrations, Auto Sea State, Auto Trim, and vessel length), that had to be set on the autopilot console in the bridge. Penn has not proven that Rhodes is even responsible for putting all of these settings into place.
Rhodes was responsible for the initial dockside setup of the device.116 However, Captain Thornton made changes of uncertain scope during the maiden voyage, in an attempt to correct the autopilot’s failure to hold course.117 The only change that is certain is Captain Thornton’s selection of the solenoid drive setting which made the autopilot inoperable.118 When Branden Rhodes was called to fix this problem, he had little reason to tamper with Auto Trim, Auto Sea State, and vessel length.
On the other hand, because he was restoring the analog drive system, Rhodes had to recalibrate the rudder feedback settings, i.e., the zero and end. points.119 On November 6, Tom Pisciotta found the rudder end points calibrated to 20 degrees.120 However, this accident occurred when the rudders went at least i0 degrees to port.121 Unlike the “rudder limit,” which only operates in certain modes, the rudder end points are akin to physical stops that the autopilot honors in all circumstances.122 Therefore, if the end points were set to 20 degrees at the time of the accident, the autopilot should have treated 20 degrees as the hard over position; it should have been unable to call for 40 or more degrees to port.
The Court understands that Penn - is claiming that loose terminal screws caused the autopilot to behave erratically in any number of ways expedient to this lawsuit. However, there is no actual support for this in the record. On the other hand, someone in Penn’s crew could have easily changed this setting and others before Pisciotta’s inspection, in an effort to troubleshoot the problem. Indeed, this, and only [518]*518this, would help explain why the analog drive calibrations that Pisciotta identified did not affect the autopilot’s operation before November 6. Rhodes need not establish that Penn actually changed these settings by a preponderance of the evidence in order to defeat liability. Because the case against Rhodes is circumstantial, Penn has the burden to rule out this alternative explanation with a fair degree of certainty, taking into account factors like control of the vessel.123 Here, Penn had “exclusive control” of the autopilot console from the time the vessel was serviced on September 15, 2010 until the accident.124 Against this backdrop, even if Penn could prove some causal link between the console settings and this accident, the Court would not find Rhodes liable.125
2. Penn has not proven that any of its alleged defects caused or contributed to this accident.
In any event, the more significant issue of law and fact in this case is whether the supposed “defects” in the autopilot identified by Penn could possibly and did in fact cause the accident that took place on November 5, 2010. Having considered the evidence thoroughly, the Court finds that Penn has not met its burden on this question.
a. None of the autopilot’s console settings could have caused the alleged malfunction.
None of the allegedly improper console settings has any apparent relationship to the accident in question. Auto Sea State allows the autopilot to navigate rougher conditions without calling for too much course correction.126 When it is engaged, the autopilot calls for less rudder, not more.127
Auto Trim allows the autopilot to do less work to compensate for wind and current that create lateral movement.128 Whereas a lateral draw would otherwise require the autopilot to give more rudder commands to one side or the other just to maintain course, Auto Trim adjusts either the port or starboard rudder angle to balance and minimize the number of commands to either side.129 However, any shift in the rudder angle caused by Auto Trim is slight; thus, it does not help explain what happened here.130
The vessel length setting facilitates Auto Trim by allowing the autopilot to calculate how far off course the vessel has gone over a given time period.131 The effect of vessel length being off by one setting on rudder [519]*519movement would be “very limited” and perhaps not even noticeable.132
The autopilot’s zero point tells gives the rudder positioning required to make the vessel go in a straight line, i.e. midships.133 If improperly calibrated, it would cause the autopilot to perform in a sloppy manner, constantly trying to compensate for the inability to hold course.134 The vessel would make an “S” around the true course heading.135 Although this could help explain an unexpected turn such as that which took place on November 5, 2010, the zero point would have had to have been set all the way to the extreme hard over position to explain what Captain Edvardsen observed. Penn has presented no evidence that this was true. Further, if the zero point were so badly miscalibrated, the issue would have surfaced before the accident.136
As explained, the end points are the rudders’ position when maximum analog drive voltage is applied.137 In this case, the end points allowed the rudders to go hard over to 40 degrees. However, they could not have caused the rudders to do so without some intervening command or signal. Thus, none of these console settings could have caused the rudders on the BLUEFIN to go hard over. All of them are essentially red herrings in this case.
b. Assuming that loose wire screws or spare cables could have caused the alleged malfunction, Penn has not proven that they did.
Even finding that Rhodes owed a duty to tighten the screws and remove the cable from the JD53 box138 and further that Rhodes failed to do so139 Penn would still need to prove that the alleged malfunction in this case was, more likely than not, caused by that failure.140 The evidence does not allow Penn to meet this burden.
The JD53 box, where Pisciotta tightened screws and removed the spare cable is the “heart and brain” of the autopilot.141 It contains the steering logic of the autopilot and connects it to other systems on board the vessel, including the EMI steering control.142 It is reasonable to infer tliat whatever happened on November 5, 2010, the JD53 box was involved, whether it was functioning properly or not.
There are three basic wiring systems in the JD53. There is the main power supply.143 Next, there is the circuit board in the analog drive that connects the autopilot to the rudder system through the EMI. 144 It controls the rudders by positive and negative voltage.145 A positive charge moves the rudders in one direction, either port or starboard; a negative charge moves them the opposite way.146 When [520]*520there is no charge, the rudders go to midships or what the autopilot thinks is midships.147 The third wiring system attaches to the motherboard, i.e., the “brains” of the autopilot.148 When its wiring is completely disconnected, it triggers an alarm on the autopilot console.149
There was reasoned expert consensus that loose wiring in the analog board could only affect the rudder feedback system by causing the rudders to return to midships.150 Therefore, loose wiring in the analog board could not have caused the accident on November 5, 2010. Similarly, Penn has presented no credible evidence demonstrating how a spare cable pressing against the analog boards might have caused the rudders to go hard over.
The evidence was murkier regarding whether loose wiring in the main power supply or the motherboard either caused or contributed to this accident. The motherboard is the brains of the autopilot and would naturally affect its responsiveness. The manufacturer’s corporate representative, Tom Burke, testified that an inconsistent power supply to the unit raised uncertain possibilities for operations.151 This would tend to credibly support Mr. Pisciotta’s instincts about loose screws and erratic behavior.152 Eric Jackson only discussed what would happen if the power supply and the wiring on the motherboard were removed altogether.153 It is hard to know the probative value of the experiments that Rhodes and Jackson did on their mock up of the autopilot and steering system, based on the limited information presented.154
In any case, Penn must do more than establish the mere possibility that loose wiring led to the malfunction that it has alleged. It must establish that loose wiring more likely than not caused the accident. Penn is unable to meet this burden for two important reasons: the delayed onset of this alleged malfunction and Penn’s failure to rule out operator error.
i. The timing of this incident undermines any possible inference of causation.
First, assuming wiring was to blame for this problem, it should have surfaced earlier and more consistently. The autopilot operated well for several months before the accident in this case. Penn only points to the BLUEFIN’s propensity to “S” around its preset course when the autopilot was engaged during its maiden voyage.155 However, these problems went away after EMI serviced the main steering system for a “starboard draw” and Branden Rhodes recalibrated the autopilot’s zero point.156 Either of these issues could have accounted for the problem. The autopilot’s response to the vessel going off course unexpectedly is to try to correct, [521]*521resulting in the “S” pattern.157 In any case, the Court cannot rule out the possibility that something besides loose wiring caused these initial problems.
A possible explanation for why there were no earlier incidents is that “excessive” vessel vibrations gradually loosened terminal screws. However, the evidence provides no support for this hypothesis. Even granting that the vessel vibrations on the BLUEFIN were severe, they never lasted long.158 Other equipment close to the engine was seemingly unaffected by whatever vibrations did occur.159
In the end, the only evidence that Penn can marshal to explain this timing issue is Tom Burke’s statement that wiring issues theoretically can cause a delayed reaction.160 Under the circumstances, this testimony is less than persuasive.
ii. Penn has not ruled out “operator error” as the lone cause of this accident.
Even giving Penn' the benefit of Tom Burke’s “anything’s possible” statements, the Court does not find causation in plaintiffs favor based on the evidence presented because Penn has not sufficiently ruled out the possibility that this accident was caused by Captain Edvardsen’s improper operation of the autopilot.161
As described above, autopilot steering on the BLUEFIN requires the harmonious operation of the EMI steering console and the autopilot together.162 The EMI console is wired into multiple steering devices, including the hand steering lever and the autopilot, and determines which device has control over the vessel. However, the EMI console does not control the autopilot.163 Changes to autopilot steering have to be made directly on the autopilot console.164
The autopilot has various operating modes, two of which — AUTO mode and AUTO-WORK mode — steer the vessel using a preset course or “heading reference” between 0 to 359 degrees.165 The “AUTO” button engages AUTO mode; the ‘WORK” button engages AUTO-WORK mode unless a non-AUTO steering mode has already been activated.166
The primary difference between AUTO and AUTO-WORK mode is that AUTO-WORK is designed to steer a vessel that is laden with cargo.167 It applies more rudder during navigational commands to compensate for additional weight.168 A second difference between the two modes is that AUTO has an off-heading alarm which sounds when the vessel is more than 10 degrees off course, whereas AUTO-WORK does not.169
Finally, according to the autopilot’s operator manual, AUTO mode gives the pilot greater ability to make course adjustments using the EMI hand steering controls. From AUTO mode, a pilot can place the [522]*522autopilot on standby (“STBY”), hand steer to a new heading using the EMI controls, and re-engage the autopilot by pressing “AUTO” or “WORK.”170 He can also hand steer to a new course heading using the EMI controls, and press “AUTO” again, without ever pressing STBY.171 Pressing “AUTO” while the autopilot is already in AUTO mode will set or “catch” a new heading reference without the need to put the autopilot in STBY.172 By contrast, from AUTO-WORK mode, this second option is not available: pressing “WORK” a second time will not reset the heading reference.173
The failure to properly input a new heading reference when switching between hand steering and the autopilot could have disastrous consequences. The autopilot would immediately navigate toward the most recent heading reference, notwithstanding the effort to steer to a new one.
Just before the accident, Captain Edvardsen was in AUTO-WORK tracking a heading of approximately 317 degrees and then hand steered to a new 359 degree heading on the EMI controls before returning to autopilot control.174 There is no question that failure to properly input the new reference heading would have caused the accident on November 5, 2010 by making the BLUEFIN sheer toward its previous 317 degree heading.175
Penn argues that it has successfully ruled out this possibility by proving that Captain Edvardsen properly operated the autopilot.176 The Captain testified that he pressed STBY on the autopilot console before switching to hand steering on the EMI controls, such that the autopilot should have “caught” a new heading reference when he pressed AUTO-WORK after returning control to the autopilot on the EMI console.177 Mate Washburn testified that he saw the Captain move for the autopilot controls, somewhere in the vicinity of the STBY button before he switched to hand steering on the EMI controls.178
Rhodes has presented four “prior inconsistent statements” in which Captain Edvardsen failed to mention pressing STBY in an effort to impeach him. The "Court agrees with Penn’s characterization of these statements for the most part.179 Captain Edvardsen’s deposition testimony, on the whole, reflects that he pressed STBY.180 Likewise, it would not have madé much sense for the Captain to go into detail about how he transferred to and from hand steering on the Penn and Coast Guard accident investigation reports because they only ask for a general explanation of how the accident happened and, at the time, he did not suspect that the sequence of buttons he pressed had anything to do with the accident.181 At the same time, this is the problem with trusting Captain Edvardsen’s memory of pressing [523]*523STBY: he had no reason to question how he had been transferring between hand and autopilot steering and whether it was correct, until he was already under pressure to remember the facts in a certain way.
In the end, the Court is unable to say with a fair degree of certainty that Captain Edvardsen did a proper heading “catch” when he brought the BLUEFIN around to 359 degrees in hand steering. The Captain’s demeanor at trial when he was testifying about this important sequence was at best shaky. Although many people get nervous when they testify, there are other factors to consider. Captain Edvardsen had only paged through the operators manual and mainly relied on the quick reference manual to operate the device.182 Despite how long he had been using the device, Captain Edvardsen was only familiar with its most basic functions.183
What most undermines the credibility of Captain Edvardsen’s testimony that he hit STBY is his confused, contradictory testimony that this step was ultimately unnecessary. On cross-examination, he testified that transferring from autopilot to hand steering on the EMI system would automatically place the autopilot console in STBY, contrary to his previous testimony and all other evidence presented.184 He also volunteered multiple times that it was possible to implement a new heading reference simply by pressing “WORK” after hand steering to a new heading.185 As already explained, there is no support for this understanding in the operator manual. Pressing “AUTO” while in AUTO mode will reset the reference heading; pressing “WORK” while in AUTO-WORK mode will not.186 On these facts, the Court cannot say with adequate certainty that Captain Edvardsen knew the correct sequence of buttons to press or that he faithfully executed that sequence at the time of the incident. This is true even without taking into account the Captain’s statements to the Coast Guard investigator.187
Captain Edvardsen’s confusion about whether “WORK” would cause the autopilot to reset its heading reference would explain why he was able to replicate the circumstances of the incident reliably for the Coast Guard and the American Bureau of Shipping. As for the “erratic” behavior observed by Tom Pisciotta, it was almost entirely consistent with the misaligned zero point that he discovered and corrected. Every time he lined up the autopilot heading with the BLUE-FIN’s actual heading and reengaged the autopilot, the rudder would shift 6-8 degrees.188 The autopilot’s zero point was misaligned by approximately that much.189 This does not constitute erratic behavior.190
As for Pisciotta’s single unsuccessful attempt to turn the rudders directly using the autopilot controls, he admitted that his knowledge of the device was imperfect [524]*524when he went to service the unit.191 As Tom Burke suggested, it is difficult to give much weight to Pisciotta’s post-accident testing without knowing precisely what he did or that he did it properly.192
For all of these reasons, Penn has not ruled out the possibility that this accident happened as the result of operator error.
II. Products Liability
Penn also claims that Rhodes is liable for selling an unreasonably dangerous product. Admiralty law incorporates products liability including strict liability for a defective product.193 The Fifth Circuit has applied Restatement (Second) of Torts, § 402A to products liability claims in the maritime context.194 Under that section, a seller may be held liable for harm caused by a defective product placed in the stream of commerce.195 To prevail on a products liability claim, a plaintiff must establish: (1) that the defendant sold the product; (2) that the product was unreasonably dangerous or defective when it left the defendant’s control; and (3) that the defect caused the plaintiffs injury.196
As Penn has indicated in its post-trial briefing, defect and causation may be inferred when an article fails in ordinary use, even if the evidence does not indicate the precise nature of the defect or the exact mechanism of causation.197 Nevertheless, Penn’s product’s liability claim must fail at in light of Penn’s failure to rule out operator error as an alternative explanation with a fair degree of certainty.198 Accordingly, Penn has no products liability claim against Rhodes.
CONCLUSION
Penn has failed to establish that any of the autopilot settings, conditions, or defects alleged in its complaint caused the accident in this case. It is at least equally probable that Captain Edvardsen failed to properly operate the autopilot on the night in question.
Accordingly,
IT IS ORDERED that judgment be entered in favor of defendant Rhodes Electronic Services Inc. and against plaintiff Penn Maritime, Inc. on Penn Maritime, Inc.’s complaint.
IT ■ IS FURTHER ORDERED that judgment be entered in favor of third-party defendants, Thoma-Sea Marine Constructors, LLC and Thomassie Properties LLC, f/k/a Thoma-Sea Shipbuilders LLC and against third-party plaintiff Rhodes Electronic Services, Inc. on Rhodes Electronic Services, Inc.’s third-party complaint.